Autobiographical Memory

Memories contribute to who we are and it is a complex system and function (Almaraz-Espinoza and
Grider, 2021)
• Memory is defined in psychology as the faculty of encoding, storing, and retrieving information
(Squire, 2009; Zlotnik & Vansintjan, 2019)
Short term memory: Information held very briefly in memory
• Digit-span task - “7 plus or minus 2” (Miller, 1956)
• Chunking concept - (Miller, 1956; Simon, 1974; Cowen, 2000)
Long term memory
Information stored in our brains for period of time, running into months or years (Eysenck, 2009)
• Differs from short-term memory with regards to storage and capacity (Cowan, 2008)
• There are types and sub-types of long term memory

Implicit memory - known as unconscious memory:

- It does not depend on our conscious recollection (Hine & Tsushima, 2018)
- It influences our physical and mental activities - the “knowing how” memory system
- Includes procedural memory, priming, emotional learning, perceptual learning and category learning

Explicit memory - known as declarative memory:

- Concerned with consciously remembering ideas
and events that have occurred through one’s life
- Its sub-types can be accessed unintentionally (semantic and episodic memory) - (Jawabri & Cascella,
2021)
- Decreases progressively with age (60)

MEMORY MODELS
THE MULTI-STORE MEMORY MODEL (ATKINSON & SHIFFRIN, 1968)
Atkinson and Shiffrin (1968) highlighted modality-specific stores
• Stimulation from the environment is received by sensory stores, some of the information is then
processed into the
short-term store, and then some information is transferred into the long-term store
CRITICAL EVALUATION
The assumption of separate stores have been influential
✓ There is evidence for modality specific stores (brain damage studies -
including the classic KF study, also Spiers et al., 2001)
✓ It has enabled researchers to expand research on the separate stores
The model is oversimplified (e.g. a single short term memory store and
only a single long term memory store)
- Learning can occur without conscious awareness and without rehearsal
- Processing order limitations exists
THE WORKING MEMORY MODEL (BADDELEY AND HITCH,
1974; BADDELEY, 1986; BADDELEY, 2001)
There has been support that short-term memory is comprised of sub-systems
✓ It reports how information is both processed and stored

✓ Evidence for separate short-term components come from brain damage studies
(KM study) and experimental research (dual-task studies)
✓ The model highlights that rehearsal is optional
It is oversimplified
- Subsystems should be further separated (Liberman, 1980)
- Questions remain regarding some components, e.g. the central executive,
and interaction between components

AUTOBIOGRAPHICAL MEMORY
Autobiographical memory encompasses our recollections of specific, personal events -
Holland & Kessinger (2010)
• The territory of autobiographical memory includes:
✓Everyday memories
✓Self-defining memories

✓Flashbulb memories
✓Memories for traumatic experiences
EVERYDAY MEMORIES
• Known as ‘real world’ memory
• It is based on how individuals use their memory to get on with their daily lives
SELF-DEFINING MEMORIES
• These are based on personal events which can evoke a range of emotions during retrieval
• These memories could be (El-Haj and Gallouj, 2019):
1. vivid
2. emotionally intense
3. repetitively recalled
4. focus on enduring concerns
5. Include unresolved conflicts

FLASH-BULB MEMORIES
• Where were you when you first heard about the September 11 attacks?
• Flashbulb memories are vivid and detailed memories of dramatic events (Brown & Kulik, 1977)
- Examples include; assassinations, natural disasters, terrorist attacks
• Features of flash-bulb memories include:
- Place
- Ongoing activity
- Informant
- Own feelings
- Others’ feelings
- Aftermath
Lead to activation of special neural mechanisms
Why are they memorable?
• Subjected to rehearsal
• They can have a major impact in our lives
• They are associated with very strong emotions
• They do not get confused with other events
• Lead to activation of special neural mechanisms
• However, they may not be as accurate as initially predicted by Brown & Kulik (1977)
Evidence
• Lack of accuracy was supported by Pezdek (2003) - asked students specific questions about the
September 11th attacks
and students had incorrect responses
• Talarico & Rubin (2003) - flashbulb memories are subject to change
• However, accuracy is related to the intensity of the emotional experience (Sharot et al., 2007) - the more
intense the
experience, the more accurate the memory
• MRI studies support the role of emotional experience in creating accurate flashbulb memories (Sharot et
al., 2007)
• Positive events can also elicit FBMs (Bohn & Berntsen, 2007)

TRAUMATIC MEMORIES
• Post-traumatic stress disorder (PTSD) is an anxiety related condition that arises from a very distressing
event
• Neuroanatomy impacted in PTSD (hippocampus - reduced volume) - Gilbertson et al., (2002); Logue et
al., (2018)
How does it relate to memory?
• PTSD results in traumatic memory formation
• In such conditions, individuals experience the traumatic event through nightmares and flashbacks
(involuntary
memories) which come through via intrusive thoughts - Reynolds & Brewin (1998)
• Cues can trigger retrieval - which disrupts cognitive functioning
• Leads to avoidance strategies
Explanations for flashbacks:
- Reapperance hypothesis (Neisser, 1967) - the same memory can appear and disappear
without being altered
- Cognitive memory explanations - differences in working memory, with greater working
memory capacity leading to better suppression of unwanted thoughts (Brewin & Beaton,
2002)
- Cognitive model explanation (Ehlers & Clark, 2000)
CHARACTERISTICS OF AUTOBIOGRAPHICAL MEMORY

Autobiographical
✓ Emphasises personal history recollection and its significance

✓ Involves stronger subjective experiences/complex memories

✓ Overlaps with episodic and semantic memory
Episodic
Emphasises events that happened at a particular time/place
- It has limitations regarding the complexity of memories
- No overlaps with semantic memory (distinctions reported in imaging data)

NEUROANATOMY OF AUTOBIOGRAPHICAL
MEMORY
The neuroanatomical basis of autobiographical memories, include:

Posterior cingulate cortex, Hippocampus, Medial prefrontal cortex, Inferior parietal lobule

FUNCTION OF AUTOBIOGRAPHICAL MEMORY

Williams, Conway, and Cohen (2008) proposed four functions of autobiographical memory:
1. Directive
• Using Past experiences to direct or guide current and future actions, thoughts and behaviours
• Helps create models (Robinson & Swanson, 1990)
• Assists in forming personal opinions and attitudes
• Buck & Gluck (2004) - Older and young adults report having learned lessons about themselves or having
gained a life philosophy from wisdom-related events
2. Social
• Autobiographical memories help initiate social bonding (Alea & Bluck, 2003)
• Facilitates social skills (understanding and empathy)
• Influences teaching/informing
• Sub-functions of the social dimension exist (Webster, 1993):
3. Self-representation
• Creating stable and enduring representation of our selves over time
• Helps maintain self-image
• Allows one to update the self while maintaining continuity (Barclay, 1996; Conway, 2003).
• Buck and Alea (2008) - individuals with a low self-concept use/recall the past for self-continuity
therefore
demonstrating that memories can be used to remind one of their identity
4. Coping with adversity
• Autobiographical memories help cope with adversity
• One can employ positive past experiences to cope with current negative experiences
• Reduced autobiographical memory specificity increases the susceptibility of depression and it is
predictive
of a more chronic course of mental health conditions (Hermans et al., 2008)

OTHER FUNCTIONS OF AUTOBIOGRAPHICAL

MEMORY
El-Mohamad & Gallouj (2019):
• Achieve long-term goals
• Anticipate events
• Contributes to emotional regulation
• Influences personality traits
• Helps transmit personal experiences
• Helps create “scripts for your lifespan” - Singer et al., 2013.

ASSESSMENT OF AUTOBIOGRAPHICAL MEMORIES

Cue recall - memory probe method (Crovitz and Shiffman, 1974):
• This is the retrieval of memory with the help of cues
• Harris and Sutton (2015) - cued recall resulted in direct autobiographical memories being recalled at a
fast rate vs.
generative autobiographical memories
• The Cue-Recalled Autobiographical Memory (CRAM) test; age related features do exist, with older adults
producing
more details memories (Gardner, Mainetti & Georgio 2015)
CRITICAL EVALUATION
Cue word results have demonstrated several reliable patterns in memory distribution over the life span:
- Retention function (Crovitz and Schiffman, 1974)
- The reminiscence bump (Fitzgerald, 1988)
- Childhood amnesia
✓ Takes into account a range of cues - verbal, visual, olfactory
What about verifying the accuracy of the memories obtained?
- Cues may lead to a rise in false memories
- Lack of experimental control/applicability can be questioned
AUTOBIOGRAPHICAL MEMORIES ACROSS THE LIFESPAN

When examining the life span distribution of autobiographical memories, three phenomena are revealed:

INFANTILE AMNESIA
• Also known as childhood amnesia
• Shown by the almost total lack of memories from the first 3 years of life; recalling very few memories
from before the age
of 3 and limited recall for events occurring between the ages of 3 and 6.
• Found in adolescence, young adults as well as the elderly
REMINISCENCE BUMP
• Surprisingly a large number of memories are remembered from the years of adolescence and early
adulthood (15-25 years)
• Increased amount of autobiographical memories from youth and early adulthood observed in adults
over 40 – Koppel and Rubin (2016)

Explanations and theories:

• Cognitive abilities due to neurological aspects (Janssen & Murr, 2008)
• Cognitive account (Pillemer, 2001) - Novelty explanations (Rubin, Rahhall and Poon, 1998) and stability
explanations (Rubin, Rahhall and Poon, 1998; Pillemer et al, 1988)
• Life narrative hypothesis explanation / narrative & identity account (Bernsten & Rubin, 2004; Gluck &
Buck, 2007)
• Life script account (Munawar et al., 2018)
RECENCY EFFECT/RETENTION FUNCTION
• The older the memories, the less likely they are to be recalled than more recent ones (Eysenck, 2009)
• We have the most autobiographical memories from events that occurred in the most recent 10-20 years
of our lives (Rubin
& Schulkind, 1997)
• Similar findings are illustrated with other cues (olfactory and visual) - Rubin, Groth, & Goldsmith (1984)
Explanations and theories:
- Normal course of forgetting
- The recency effect
- Bias and preference theory

INDIVIDUAL DIFFERENCES IN AUTOBIOGRAPHICAL

MEMORIES
• Gender
Females
- Record more childhood memories (Davis, 1999)
- Have faster accessibility (Davis, 1999)
- Report more vivid experiences and emotion (Grysman & Hudson, 2013)
• Personality
Openness
- increased levels = greater vividness and emotion of AM experiences (Rubin & Siegler, 2004)
- correlates positively with directive and self function aspects of AM (Ramassen and Berntsen, 2010)

Personality continued
Neuroticism
- correlates positively with self function aspects of AM (Ramassen & Berntsen, 2010)
Remaining big 5
- less clear roles (Ramassen & Berntsen, 2010)
• Culture
- Western children and adults often recall more detailed, more specific, and more self- focused memories
- Euro-American adults are able to access more distant/more detailed very-long- term memories Wang,
2001, 2006;
Wang & Conway, 2004).

Culture
- Han et al., (1998) - provides evidence for cultural differences in AM
- Both European and Asian American adults recalled memories that focused less on social interactions
(Wang, 2011)
AUTOBIOGRAPHICAL MEMORY - THEORETICAL VIEWS
Self Memory System Model - Conway & Pleydell-Pearce (2000)
Autobiographical memory depends on:
1) Autobiographical memory base
- Lifetime periods
- General events
- Event specific knowledge
2. The working self
- The self
- Influence the memories stored within the autobiographical knowledge base
• According to the theory, autobiographical memories can be accessed through generative or direct
retrieval
Self Memory System Model (further developed) - Conway (2005)
• Event specific knowledge is now divided into the
conceptual-self and episodic memories
• This is a hierarchical structure of an overall life story,
including
- Broad themes
- Lifetime periods
- General events
- Episodic memories
CRITICAL EVALUATION
Support for:
- Major goals influence our autobiographical memory (Wioke et al., 1999; Newman & Lindsay,
2009)
- Autobiographical memory has a hierarchical structure - brain damaged patients (Conway &
Rubin, 1993)
- Autobiographical memories and the self are closely connected (Conway & Holmes, 2004)
✓ Neural evidence for the distinctions between the retrieval types as stated in the model (Addis et al.,
2012)
AUTOBIOGRAPHICAL MEMORY DISTURBANCES
• Mental health
- Depression:
Diminished (and slower) access to positively experienced autobiographical past events (Dalgeish &
Wener-Seidler, 2014)
Do not experience mood improvements following recollection of positive autobiographical memories
(Joormann, Siemer & Gotlib, 2007)
Less vivid and emotionally intense recall compared to controls ( Wener-Seidler & Moulds, 2012)
There have been mixed findings - (Lui et al., 2013)
- Obsessive compulsive disorder:
OCD patients have difficulty retrieving autobiographical memories Vs. Healthy control group (Wilhelm et
al., 1997)
OCD patients showed longer retrieval latencies when retrieving memories (Wilhelm et al., 1997)
May reflect co-morbid symptoms rather than OCD per se
Mental health
- Boderline personality disorder (BPD):
BPD patients demonstrate impaired autobiographical memory functioning (Beran, Richman & Unoka,
2018)
Recall greater negative life events compared to healthy controls (Renneberg et al., 2005)
Autobiographical memories emphasise rejection (Rosenbach & Renneberg, 2015)
• Cognitive conditions; Amnesia
- Dissociative Fuge
Loss of autobiographical memories; Patient NN - Hennig-Fast et al., (2008)
Severe loss and uniform loss of memories for facts and events across all time period (Harrison et al.,
2017)
Neuropsychiatrics conditions
- Confabulation
Generating false memories without intending to deceit (Wiggins & Bunin, 2020)
Associated with frontal lobe diseases (AD and Korsakoff’s Syndrome) and can also effect the average
person too
Impacts autobiographical memories the most (Barba & Denes 1990)
Types include: provoked type & spontaneous type

UNIT: 2: Object Perception

WHAT IS OBJECT PERCEPTION/RECOGNITION

A type of visual perception that we do thousands of times everyday!
• “Object perception or object recognition is the process in which visual input is assigned a meaningful
interpretation that is available to perceptual awareness” - (p.1, Grill-Spector, 2008).
It is complex:
- Objects in the environment overlap
- Objects in the same category vary in their property
- We can recognise objects over a wide range of viewing distances
VISUAL PROCESSING
Light hits the eyes and passes through the cornea and lens creating an image on the retina that is
reversed.
The information from the retina is sent via the optic nerves and passes through the optic chiasm.
The axons from the outer halves of each retina travel to the hemisphere on the same side of the brain,
whilst those from the inner halves of the retinas travel to the opposite hemisphere.
Therefore, each side of the visual space is represented in the opposite brain hemisphere.
The sensory input is then relayed by the lateral geniculate nucleus and transmitted to path the primary
visual cortex

Dorsal stream:
- Stretches from the primary visual cortex (V1) in the occipital lobe and flows to the parietal lobe
- Guides actions by recognising where objects are in space
- Known as the ‘Where’ pathway
Ventral stream:
- Stretches from the primary visual cortex (V1) to the inferior temporal lobe
- It is involved in object identity (Sheth & Young, 2016)
- Known as the ‘what’ pathway
- Plays a crucial role in processing
EVIDENCE FOR THE TWO STREAMS AND THEIR ROLES

Brain damaged patients

- Humans with right temporal lesions have impaired ability to perform perceptual tasks (Kimura, D.,
1963)
- Those with dorsal stream damage can describe object size, location and orientation but are unable to
use their vision to control object grasping movements (Goodale, 2005)
- Patients with damage to the ventral stream are unable to perceive the size, shape, and orientation of
objects (Goodale, 2005)
Neuroimaging findings
- Using vision to recognize objects results in selective activation in ventral stream areas such as the lateral
occipital area and the fusiform gyrus (Grill-Spector, 2003)
- Using vision to reach out and pick up objects results in no longer showing selective activation in these
ventral stream areas, but instead show activation in dorsal stream regions, such as the anterior part of the
intraparietal sulcus (Culham, 2004)

PROCESSING DIRECTIONS
• Top-down processing
• Bottom-up processing
Bottom-up processing begins with the retrieval of sensory information from our external environment to
build perceptions based on the current input of sensory information. Top-down processing is the
interpretation of incoming information based on prior knowledge, experiences, and expectations.

PATTERN RECOGNITION
A pattern is an object
• During recognition, given objects are assigned to prescribed classes
• Pattern recognition is the ability to identify two-dimensional patterns (e.g. letters, symbols)
• It is also the process of matching perceptual info with long term memory info
• Includes serial processing and parallel processing
• Our perceptual system is flexible

PATTERN RECOGNITION - THEORETICAL EXPLANATIONS

Feature based theory -
- Patterns consisting of a set of specific features or attributes (Jain & Duin, 2004)
- Different stimuli have combinations of features, for example, the letter T is made up of

Feature based theory -

- Feature processing happens before global processing
- Overall, we make discriminations based on a characteristics of stimuli
• Some points for evaluation:
It has been argued that feature components are stored in memory. There have been contrasting evidence
for whether feature processing can happen before global processing and vice versa (Dalrymple et al.,
2009; Navon, 1977)

PERCEPTUAL ORGANISATION
• This is a fundamental issue in visual perception
• Eysenck (2008) noted that it is based on our ability to accurately work out:
1. Which parts of visual information presented to us belong together and thus form objects
2. Which parts should be nearer and which further away
• Based on the process of grouping visual elements to form a whole
• Studied by the Gestaltist’s (Koffka, Kohler & Wertheimer) who led the Gestalt Psychology movement,
and their primary principle for perceptual organisation was the law of Pragnanz.
LAW OF PRAGNANZ
• Law of Pragnanz: “Of several geometrically possible organisations that one will actually occur which
possesses the best, simplest and most stable shape” - Koffka, 1953, p.138
• Humans will naturally perceive the simplest organisation of their visual environment, which is a whole
pattern, rather than the individual elements (Law of Prä gnanz).
• There are predictable ways in which our brain tries to perceive a whole, these have been translated into
a set of laws/ principles to explain perceptual organization

GESTALT LAWS/PRINCIPLES OF PERCEPTUAL ORGANISATION

The law of proximity Visual elements close together tend to be grouped together
• Horizontal vs. Vertical dots
The law of similarity: • Similar elements tend be grouped together
The law of good continuation: We tend to group together elements requiring the fewest changes or
interruptions in straight or smoothly curving lines
The law of closure: Missing parts of a figure get filled in, in order to complete it.
The figure-ground segregation: One part of the visual field is identified as a figure, and the rest is treated
as less important and so forms the background
• Figure = distinct form and shape
• background = lacks form
CRITICAL EVALUATION
Nearly all the laws of grouping are support within the literature - Elder & Goldberg (2002)
✓ There is evidence for the figure-ground law. It occurs in absence of attention, but only with simple
objects -Kimchi & Peterson (2008)
✓ Neuronal support for figure-ground segregation - Layton et al., 2012
Questions about applicability as Gestaltist studied artificial figures mostly
- New laws? - Geisler et al., (2001)
- How and why does it happen, and what about conflict between different
laws of grouping? - Quinlan and Wilton (1998); Eysenck (2008)
- Learning and past experiences were not taken into account - Bhatt & Quinn
(2011); (Barense et al., 2012).

FROM 2D TO 3D - SPATIAL FREQUENCY

• Visual processing starts with coarse-to-fine processing.
Different cells specialised in responding to high or low spatial frequencies in the primary visual cortex.
Low spatial frequencies in visual input travel rapidly to higher-order regions via the magnocellular
pathway.
High spatial frequencies in visual input travel more slowly via the parvocelluar pathway.
The speed difference is the reason why coarse processing normally precedes fine processing.
However, coarse-to-fine sequencing is flexible-depends on the task demands

THEORETICAL APPROACHES TO OBJECT

RECOGNITION
MARR’S (1982) COMPUTATIONAL APPROACH TO VISION
• Object recognition involves a series of processing stages
• Observers construct series of representations that provide increasingly detailed info about the
environment, and these stages include the following:
1. Primary sketch: A 2D outline of light-intensity changes, including information about edges, contours
and blobs
2. 2½D sketch: Includes information about surfaces (e.g. depth and orientation), making use of
information from shading, texture, motion, binocular disparity, etc. It is object-centred or viewpoint-
dependent
3. 3D model representation: Describes shapes of objects, and their positions. It is viewpoint-invariant.
Match representation with semantic memory

VIEWPOINT DEPENDENT & VIEWPOINT INVARIANT

Viewpoint-dependent theories of object recognition:

• Perception depends on the structures associated with the observer
• Object recognition is faster and easier when objects are seen from certain angles
• Known as observer-centred because the recognition of the object depends on the angle/orientation the
viewer is looking at the object from.
• Viewpoint-invariant theories of object recognition –
• Perception is based on the structures independent from the observer and these structures are universal
and invariant
• you recognise that the object is stable, regardless of the information at the retina
• Object recognition is equally rapid and easy regardless of the angle
CRITICAL EVALUATION
The theory has been very influential - Glennerster, 2007
✓ Its a comprehensive computational model that inspired others

✓ The model distinguished between viewpoint-dependent and view-point invariant elements

Mostly bottom-up processing, and ignores top-down processing
- Ignored the role of past experience and prior knowledge
- 3d models are too complex (Mather, 2015)

BIEDERMAN’S RECOGNITION-BY-COMPONENTS (RBC) THEORY

Geons
These are geometric ions
- Examples of geons are blocks, cylinders, spheres, arcs and wedges
- Objects are made up of 36 different geons that make up the building block of all objects
- Geons can be arranged in almost endless different ways. Object recognition depends primarily on the
identification of geons
• Perceived geons are compared with objects in our stored memory to identify the object we are seeing
The first step of object recognition is edge extraction: determine the edge
• This is followed by determining the concavities of an object:
- Where do lines and edges join?
- The area where two edges meet
• Non-accidental properties are then taken into account: Exploring whether regularities in the visual
image reflect actual or non-accidental regularities in the real world, rather than accidental characteristics
of a given viewpoint.
• These help determine the components, which are then matched to object representations we have
The RBC theory is an example of the view-point invariant approach
• Object recognition in this theory is typically viewpoint-invariant, because there are 5 invariant
properties of the edges of geons.
- Curvature, Parallel Cotermination, Symmetry and asymmetry, Collinearity
• Since an object’s geons can be identified from numerous view points, object recognition should not
depend on viewpoint provided that all of an object’s geons are visible. We are able to recognise objects
when they are partially hidden, but only when certain conditions are present, Biederman (1987):
1. The invariant properties (e.g., curvature, parallel lines) can still be detected even when only parts of
edges can be seen
2. As long as the concavities of a contour are visible, the missing parts of a contour can be restored
3. There is normally enough information available for recognising complex objects, and so they can still be
identified when some of the geons are missing
CRITICAL EVALUATION
Evidence for:
- Concavities are important, which is supported by from eye movement studies (Leek et al., 2012)
- Edge information is more important than surface information (e.g. colour) - Biederman (1987)
- Neuroscientific studies demonstrate the importance of geons for object recognition (Vogels et al., 2001)
✓ Influential overall
Researchers have argued that object recognition is view-point dependent - Tarr & Bulthoff (1995)
- There have been questions regarding the importance of edge information - Sanocki et al., (1998)
- Animal studies have shown that surface cues are more important than edge cues - Parron and Washburn
(2014)
- Focuses on bottom-up processing
TOP DOWN PROCESSES
• Backward processing neurons are associated with top-down processing - Gilbert & Li (2013)
• Top-down processing occurs when:
1. Object is ambiguous - Goolkasian & Woodberry (2010): primes influence the
interpretation of the ambiguous image
2. Object is difficult to recognise - Bar et al., (2006)
3. Visual input is ambiguous - Bar & Ullman (1996), in such cases the contextual
surrounding will plays a role in interpretation
Mood also influences the speed of object recognition - Panicello et al., (2016)
DISORDERS OF OBJECT RECOGNITION
OBJECT RECOGNITION DISTURBANCES
• Stroke
- This occurs when the brain is starved from blood supply, and there are different types including
ischemic stroke, hemorrahagic stroke and a transient ischemic attack
- Damage to the right hemisphere impairs recognition of objects shown at an unusual angle - Layman &
Greene (1988)
- Patients are unable to match unfamiliar objects which had been rotated - Layman & Greene (1988)
- Case studies: - Patient J.S: inability to “see” objects, but identification of objects by their characteristic
smell or taste was possible. In neuropsychological testing- J.S. recognized only 3 of 12 real objects, and
tactile exploration of the visually unrecognized objects led to correct identification in 5 objects - Karnath
et al., (2009)
Agnosia
- Apperceptive form
- Caused by impaired visual processing that results in the poor perception of an object (Kim, 2016)
- Result in misperception of shapes due to defects in representing the elementary properties of curvature,
surface and volume
- Failure integrating multiple elements into a perceptual whole
- Aieteology: severe occipital lesions and residual field defects
- Associative form
- Caused by disorders that affect the associative cortex (Kim, 2016)
- Results in the correctly formed visual percepts being poorly matched with previously processed sensory
data and recognition (Kim, 2016)
- Inability to access the stored knowledge related to the percept (Berti & Neppi-Modon, 2012).

UNIT 3: Expertise and Problem Solving

DISTINGUISHING THE DIFFERENCES (DEFINITIONS)
• Problem solving:
- Based on moving from the recognition that there is a problem through a series of steps to the solution
- Associated with how we solve problems that don’t require specialist knowledge
Expertise:
- High level of domain-specific knowledge and skills accumulated with age or experience
- Associated with how one becomes an expert and the different knowledge and skills experts use
compared to novices

Points to consider:
- There is a lot of overlap between problem solving and expertise
- Knowledge seems to be more important in research in expertise, rather than problem solving. And
individual differences seem to be of interest in the expertise research, rather than problem solving
- Expertise research mostly focuses on what the differences are between experts and novices regarding
their knowledge and problem solving skills

PROBLEM SOLVING
• Problem solving is vital because it is “a crossroads, where many different processes come together in
the service of the needs and goals of an individual” (Weisberg, 2018, p. 607).
• There are 3 major aspects to problem solving:
1) it is purposeful
2) It involves controlled processes; doesn’t always rely on “automatic” processes.
3) A problem exists when someone lacks the relevant knowledge to produce an immediate solution.

There are two types of problems (Eysenck & Keane, 2020):

1) Well defined - these are clearly specified, and goals/methods available for solving are clearly evident
2) Ill defined - these are underspecified, and goals/methods available for solving the problem may be
unclear (Goel, 2011)
• There is also a differentiation between:
1) Knowledge-rich problems
2) Knowledge-lean problems
BRAIN REGIONS INVOLVED IN PROBLEM SOLVING
• There is a ‘core problem solving network’ involved with regards to three types of problems;
mathematical problem solving, verbal problem solving and visuospatial problem solving, and there is a
conjunction activated region too, Bartley et al., (2018).
• Creative problem solving is associated with strong activation in frontal regions of the left hemisphere -
Fink et al., (2009)

THE GESTALT APPROACH - INSIGHT

• Gestaltists distinguished between:
1) Reproductive thinking - involves the systematic reuse of previous experiences
2) Productive thinking - involves novel restructuring of the problem and is more complex
• Problems requiring productive thinking are often solved using ‘insight’
• Insight is the experiences of suddenly realising how to solve the problem - i.e. an “Ah-ha” moment
(Kounios & Beeman, 2014).

An example of an insight problem is the mutilated draftboard (or chessboard, or checkerboard) problem:
INSIGHT - EVIDENCE
The research surrounding insight:
• Metcalfe and Wiebe (1987) – measured feelings of “warmth” in insight and non-insight problems.
• Kounios & Beeman (2014) – insight associated with anterior cingulate cortex and pre-frontal cortex.
• Ellis at al. (2011) – used an eye tracker to measure insight during an anagram task.

FACTORS FACILITATING INSIGHT

HINTS
• Problem-solving can be facilitated with hints, which provide information about given problems (Nakai
et al., 2017)
• Hints increase the number of solutions produced (Eysenck & Keane, 2020)
• Even subtle hints are useful (e.g. Maier’s, 1931, Pendulum Problem)
• Thomas & Lleras (2009) showed that hints can be effective without conscious awareness of their task
relevance.
2. INCUBATION AND SLEEP
Incubation is a a stage of problem solving. Furthermore, during incubation, the problem is put to one side
for some time (Eysenck & Keane, 2020)
• Wallas (1926) claimed problem solving can benefit from incubation , which “arises when the solution . . .
comes to mind after a temporary shift of attention to another domain” (Sio & Ormerod, 2015, p. 113).
• Wallas (1926) argued the subconscious continues to work towards a solution during incubation.

“Sleeping on a problem” is effective form of incubation (Wagner et al., 2004)

• Simon (1966) argued that incubation is important because it allows us to forget strategies that were
misleading in the first place
• Penaloza & Calvillo (2012) found that forgetting misleading information is important. Past experience
increases our ability to solve problems.

PAST EXPERIENCES AND PROBLEM SOLVING

• Past experience increases our ability to solve problems. However, Gestaltists argued that this is not the
case. The negative effects of past experience are demonstrated by what is known as functional fixedness.
This refers to the inflexible focus on the usual functions of an object in problem solving.
- Supported by Duncker (1945)
- Supported by Ye et al. (2009)
Challoner (2009) functional fixedness could be solved by noticing a feature that isn’t noticed normally
noticed and then form a solution based on that obscure feature.
The mental set:
• This involves continuing to use a previously successful problem-solving strategy even when
inappropriate or suboptimal.
• In other words, it is the tendency to use a familiar problem-solving strategy that was successful in the
past, but might not be appropriate now.
• Bilalic et al. (2008) used a chess example to demonstrate this

PROBLEM SOLVING STRATEGIES

• Newell and Simon (1972) argued that it is possible to produce systematic computer simulations of
human
problem solving. General problem solver designed, and assumptions include:
1. Information processing is serial (one process at a time)
2. People possess limited short-term memory capacity.
3. Relevant information from long-term memory can be retrieved.
• Newell and Simon (1972) reported that humans generally use rules of thumb (heuristics) when trying
to solve problems.

Some heuristics and strategies that humans use during problem solving include:
1. Means-end analysis (Newell & Simon, 1972) - based on creating a subgoal to reduce the difference
between the
current state and the goal state.
2. Hill climbing (Newell & Simon, 1972) - focus on making moves that bring you closer to the goal
3. Meta-reasoning - monitoring processes that influence the time, effort and strategies used during
reasoning and
problem solving.
4. Progress monitoring (McGregor et al., 2001) - enables one to change their strategy if needed
5. Planning is also critically involved in problem solving (Goel and Grafman, 1995)

HOW DO COGNITIVE CONDITIONS IMPACT PROBLEM SOLVING

Dementia
• This is the decline in memory and other thinking skills that result in reducing a person's ability to
perform
everyday activities (Emmady & Tadi, 2021)
• Memory impairment is followed by impairment in problem-solving (Kumar et al., 2020)
• Studies have demonstrated disorders of executive function in dementia and Alzhemier's disease
(Rabinovici et al., 2015; Matyr & Clare, 2012)
Autistic Spectrum Disorders (ASD)
• Many individuals with an ASD have significant cognitive impairments, including defitcits in problem
solving
• On the twenty questions task (TQT) high-functioning children and adults with ASD take more guesses
on the game (Minshew et al., 1994)
• However, savant characteristics should be considered (Hughes et al., 2018)
HOW DO PSYCHIATRIC CONDITIONS IMPACT
PROBLEM SOLVING
Depression
• In major depressive disorder (MDD), deficits have been shown on tests measuring problem solving and
planning (Naismith et al., 2003). Depression impacts everyday problem-solving ability (Yen et al., 2018)
PTSD
• Higher PTSD symptom scores predicted poorer problem solving skills - Kasckow et al., (2011). Problem
solving therapy has been implemented in those with PTSD as well as those with depression to improve
their symptoms (Alexopoulous et al., 2003)
OCD
• Patients with obsessive-compulsive disorder (OCD) demonstrate deficient planning capacity in the
Tower of London (TOL) problem solving task - Bey et al., (2018). Abnormalities in planning/problem
solving are shared features in OCD patients, and might be trait markers for OCD susceptibility - Bora
(2020)

EXPERTISE
• In expertise, knowledge and skills are developed over years of practice and people become experts in
their field
• Expertise involves extremely high level thinking
• Research on expertise has mainly focused on:
1) Chess experts
2) Medical experts
• Expertise is usually assessed using knowledge-rich problems
CHESS EXPERTISE
• It takes 10 years of intense practise to become a chess master - Chase & Simon (1973). Chess masters
recall the chess board positions much more accurately than less expert players (De Groot, 1965).
The template theory (Gobet & Waters, 2003)
• This theory proposes the notion that chunks that are used frequently develop into more complex data
structures known as templates:
- These are schematic structures consisting of a core and slots. They typically store information relating
to ten pieces or more. They are flexible and adaptable due to slots.

Predictions and evidence for the template theory:

• Prediction 1: The chances are larger and fewer in number than assumed by chunking theory. This is
supported by Golbet and Clarkson (2004)
• Prediction 2: Good chess players rely on their template-based knowledge, rather than use of slow,
strategybased processes. Charness et al. (2001) showed that expert players assess positions very quickly.
Moxley et al. (2012) found slow strategy-based processes important.
Prediction 3: Expert chess players store the precise board locations of pieces. Further explored by
McGregor & Howes (2002)
• Prediction 4: Expert chess players have better recall of random chess positions. Gobet & Simon (1996)
and Gobet & Waters (2003) – experts were better than novices in recalling random chess positions.

Predictions are mostly support

✓ Experts do appear to store board position information in a few large templates

✓ The reason why expert chess players can complete games of chess very quickly is because of the
superior template knowledge they have and they also have better recall for random positions, than non-
experts.
Slow strategic search processes are used more by expert players than the theory assumes
- The precise nature of what is stored in long-term memory for chess experts is controversial
- Gregor and Howes (2002) showed that the attack and defence relationships between pieces may be
more
Important

MEDICAL EXPERTISE
• Medical experts use implicit reasoning; novices use explicit reasoning (Engel, 2008). Eye-tracking good
for determining the focus of attention while examining case slides. Krupinski et al. (2013) - found there
was a significant reduction in the number of fixations per slide and there was less examinations of non-
diagnostic regions. Kundel et al. (2007) – tracked eye movements during examination of mammograms
and found fast fixation was excellent predictor of high performance.
Gegenfurtner et al. (2011) identified several differences between experts and non-experts that were
common across domains:
- shorter fixations, faster first fixations on task-relevant information, more fixations on task-relevant
information, fewer fixations on task-irrelevant areas, longer saccades (rapid eye movements).
• Supports the information-reduction hypothesis (Haider & Frensch, 1999) and the holistic model
(Kundel et al., 2007).
• Melo et al. (2012) suggested that experts use visual strategy, whilst non-experts use a more analytical
approach.

THEORY OF DELIBERATE PRACTICE

• Expertise can be developed through deliberate practice (Ericsson and Towne, 2010)
• The four aspects of deliberate practice includes:
1. The task is at an appropriate level of difficulty
2. The learner is given informative feedback about performance.
3. The learner has chances to repeat the task.
4. The learner has the opportunity to correct his/her errors.
• Ericsson and Kintsch (1995) proposed the notion of long-term working memory: experts store relevant
information in longterm memory that can be readily accessed through retrieval cues in working memory
• Two key predictions - 1) acquisition of expertise depends on amount of deliberate practice, not just
number of hours devoted to practice; 2) deliberate practice is all that is needed to develop expertise.

EVIDENCE FOR DELIBERATE PRACTICE

• Specific case studies - Ericsson and Chase (1982) studied SF, who increased his digit-span from 7 digits
to 80 digits after two years by making use of long-term working memory.
• Neuroimaging studies - Guida et al., (2012, 2013) reviewed neuroimaging findings in studies of experts
performing tasks involving use of working memory.
• Campitelli & Gobet (2011) found that performance level of experts is highly correlated with amount of
deliberate practice.
• Tuffiash and colleagues (2007) found that elite Scrabble players also engaged in other deliberate
practice activities.

LIMITATIONS FOR DELIBERATE PRACTICE

• Very difficult to measure. Although expertise is strongly associated with amount of deliberate practice, it
does not suggest a causal effect.
• The evidence suggests deliberate practice is necessary (but not sufficient) to produce high levels of
expertise (e.g. Howard, 2009; Hambrick et al., 2014)
• The theory ignores innate talent
UNIT 4: Language comprehension and reading
WHAT IS LANGUAGE
Language is a fundamental skill!
• The American Psychological Association (2020) outlined some definitions of language as:
1. A system for expressing or communicating thoughts and feelings through speech sounds or written
symbols
2. A communicative system used by speakers, with distinctive vocabulary, grammar, and phonological
systems
• Language is unique to humans; other species do communicate but they have functions for a particular
context
APPROACHES AND FUNCTIONS OF LANGUAGE
Crystal (1997) reported that there are 8 different functions of human language. Different approaches
within the field of psychology highlight the different functions of language, including:
• Skinner (1957) – behaviourist approach to language development (language is reinforced through
conditioned responses)
Chomsky (1957, 1959) – Theory of Universal Grammar (we are all born with the ability to learn the rules
of language).

READING
Reading is an important skill, and it is therefore important for us to understand the processes involved in
reading (Eysenck & Keane, 2020), which can include:
- Perceptual processes
- Cognitive processes
• It is important to note, that reading can sometimes be more difficult than speech perception because:
1) Reading lacks prosodic cues (pitch, intonation, stress and timing)
2) Reading lacks body language and gestures

Processes and structures involved in reading comprehension include (Perfetti & Stafura, 2014)
1. Various kinds of stored information;
- Lexicon
- Orthograpic knowledge
- Knowledge about the world
- Linguistic knowledge
Processes and structures involved in reading comprehension include (Perfetti & Stafura, 2014)

2. Word identification followed by text comprehension;

- Exploring the parser, Drawing inferences, - Producing a situation model representation (i.e. an
integrated mental representation)
3. There is flexibility in the order by which reading processes occur, but overall, reading involves complex
processes
METHODS
Lexical decision task: involves rapidly deciding if a string of letters forms a word.
• Naming task: involves saying a printed word out loud as quickly as possible.
• Eye-tracker: records eye-movements and measures attention-focused
• Priming task: a prime word is present shortly before the target word – both are related (e.g. in spelling,
meaning or sound) (Rastle & Brysbaert, 2006):

EEG methods and time-locked EEG activity (i.e. ERPs - event related potentials): monitoring brain activity
(Rabovsky et al., 2018)
READING - WORD RECOGNITION
The Processes involved in word recognition:
The Interactive Activation Model - proposed by McClelland & Rumelhart, (1981):
• Bottom up and top down processes interact to recognise words
• It consists of a feature level, letter level and word level
• This model has been very influential (Harley, 2013)
• The model accounts for effects including:
1. word superiority effect
2. effects of orthographic neighbours

WORD SUPERIORITY EFFECT

Word superiority effect (Reicher, 1969) highlights that context is important:

ORTHOGRAPHIC NEIGHBOURS EFFECT

Orthographic neighbours: With reference to the target word, the number of words that can be formed by
changing one of its letters.
• Large neighbourhood word: CAT (neighbours include: MAT, SAT, FAT)
• Small neighbourhood word: Hippopotamus (no neighbours)
• Orthographic neighbours influence recognition time.
• Orthographic neighbours facilitate word recognition, if they are less frequent in the language than the
target word itself. However, they have an inhibitory effect if they are more frequent than the target word
(Chen & Mirman, 2012).
CRITICAL EVALUATION
The theory has been very influential (Harley, 2013)
✓ Good example of how connectionist processing system can be applied to visual word recognition

✓ The model explains the word superiority effect and orthographic neighbours effect. What about
semantics? (Meyer & Schvaneveldt, 1971)
- What about phonological processing? (Rastle and Brysbaert, 2006)
- It does not account for the role of context (Kutas et al., 2011)
- Puts too much importance on letter order
- The model only accounts for 4 letter words

READING DEFICITS
DYSLEXIA
- Also known as a reading disability (Siegel, 2016)
- Causes impairments in speed and accuracy of word decoding
- Impacted comprehension
- It is a stable condition (Bruck, 1990)
Explanations:
• Visual difficulties or problems with hand-eye coordination (Orton, 1925)
• Dual route theory (Colthart et al., 1993) - Most dyslexics have difficulty with phonological route because
they lack phonological skills:
- Supported by Lundberg, Frost & Peterson (1998)
- Supported by Elbro & Petersen (2004)
Genetics play a role (Pennington, 1991; Seigel, 2006):
- Chromosomes 6
- Chromosome 15
SPEECH PERCEPTION
Most important form of auditory perception, and differs from other kinds of auditory perception (left-
hemisphere dominance).
• “The process in which a listener decodes, combines, and converts an incoming stream of otherwise
meaningless sound created by the speech production process into a meaningful sequence and
phonological representation” - APA, (2020)
• Speech perception can sometimes be more difficult than reading because:
- Reading can be done at own pace, whereas some people speak fast
- In reading words are spaced out; in speech hard to tell when one word ends and the other begins
- Background noises can be distracting in speech
- Words no longer available in speech – so demands are greater on STM
• Speech perception involves several stages:
1) Decoding speech signals
2) Identification of phonemes or syllables
3) Word identification
4) Interpretation
5) Integration with context
COPING WITH LISTENING PROBLEMS
Understanding speech not very straightforward (Mattys et al., 2012) Ways we cope:
1. Segmentation:
- Dividing the almost continuous sounds of speech into separate phonemes and words
- Relies on top-down and bottom-up cues
2. Lip reading:
- The McGurk Effect: auditory illusion that demonstrates interaction between hearing and vision in
speech perception (McGurk & McDonald, 1976)
- McGurk Effect occurs when crucial word presented in semantically congruent sentence (Windmann,
2004)
SPEECH PERCEPTION DISORDER
Auditory processing disorder
- These individuals have poor recognition of speech in noise (Lagace, Jutras & Gagne, 2010)
- Difficulty in complex listening situations
- Difficulty comprehending verbal instructions
- It is not a hearing problem (NHS, 2020)
• Auditory agnosia
- Impairments in sound perception despite intact hearing (Slevc & Shell, 2015)
- It can affect all types of sound perception or it can be specific to a particular domain
- Different types of therapy has been used in treatment, e.g. speech therapy and training with musical
sounds (Kim et al., 2018)

LANGUAGE COMPREHENSION
Higher level processes involved in language comprehension (e.g. sentences, stories) are generally the
same whether you are reading or listening.
• There are two main levels of analysis
1. Syntactical structure of a sentence (i.e. parsing):
- Syntax: involves the formation of sentences and pays close attention to the word order
- Grammar: concerned with the way in which words are combined, but also includes punctuation, etc
2. Meaning of sentence (i.e. semantics)
- Pragmatics: the study of intended meaning
- Sometimes the literal meaning is not intended (e.g. irony, sarcasm, and understatement)
PARSING
Parsing is analysis of the syntactical or grammatical structure of sentences.
• Research into parsing looks at the relationship between syntactic and semantic analysis:
- Syntactic analysis precedes semantic analysis
- Semantic analysis occurs prior to syntactic analysis
- Syntactic and semantic analysis occur simultaneously
- Syntax and semantics are very closely associated
• A major problem with research on parsing is that it has been based on the English language, which
methodologically impacts research findings
• Ambiguous sentences are used to study parsing.

PROSODIC CUES
Prosodic cues include stress, intonation, pauses and duration
• Prosodic cues are used by listeners to work out the grammatical structure of speech
- Example: John, go to the library for me (Hirotani et al., 2006)
• Studies have demonstrated that reading silently also involves prosodic cues (Steinhaur & Friederici,
2001)
PARSING MODELS
There are different models of parsing
1. Two-stage, serial processing theories
- Parser only entertains one syntactic structure at a time
2. One-stage, parallel processing theories
- Parser entertains multiple syntactic structures at a time
GARDEN PATH MODEL - FRAZIER & RAYNER (1982)
Readers and listeners can be “led up the garden path” (i.e. misled), by ambiguous sentences
• An example - “The horse raced past the barn fell”:
- This can be tricky to understand
- We already concluded that “the horse raced past the barn” was a complete sentence, then when we read
“fell” it didn’t make sense.
• The model assumes that meanings of words do not influence the initial assignment of grammatical
structure, only one meaning considered at a time and simplest syntactical structure chosen first.

CRITICAL EVALUATION
There is support for the principle of minimal attachment (Fraizer & Rayner, 1982):
- the girl knew the answer by heart
- The girl knew the answer was wrong
✓ Evidence that readers’ use of late closure was shown by van Gompel and Pickering (2001). Example:
“After the child had sneezed the doctor prescribed a course of injections”.
✓ Semantic dementia didn’t affect detecting grammatical violations (Bredin & Saffran, 1999)
Visual context is important in influencing initial parsing (Spivey et al., 2002)
- Meaning of words does influence the initial assignment of grammatical structure (Trueswell et al., 1994)
- Pauker et al. (2012) found that misleading prosody can prevent principles of minimal attachment and
late closure:
- “When a bear is approaching the people, the dogs come running” – coincides with sentence structure
- “When a bear is approaching, the people, the dogs come running” – doesn’t coincide with sentence
structure and contains misleading pauses
- Here, participants accepted meaning of 1 over 2
CONSTRAINT-BASED MODEL - MCDONALD ET AL., (1984)
All sources of information or constraints are available immediately to someone processing a sentence
(e.g.
syntactic, semantic, general world knowledge). These are called constraints as they limit, or constrain, the
number of possible interpretations
• The model assumes that all relevant information is immediately available to the parser
• The model assumes parallel processing
The model assumes flexibility in parsing decisions, because the information we attend to depends on past
linguistic experience
• According to the theory, the processing system uses four language characteristics to resolve sentence
ambiguities:
1. Grammatical knowledge constrains interpretations
2. Information associated with a word is typically not independent
3. A word may be more ambiguous in some ways than in others
4. Interpretations generally differ in probability on the basis of past experience

This account is a development of the previous constraint-based model

• The account assumes speakers use various strategies to reduce processing demands:
- Start with common words and syntactically simple phrases while the rest of the sentence is planned
- Favour more practised and easy sentence plans
• Many constraints used by listeners to facilitate comprehension depend on their knowledge of language
strategies used by speakers.

CRITICAL EVALUATION

More efficient to use all relevant information. Word meaning (Trueswell et al., 1994) and context (Spivey
et al., 2002) are used.
✓ Evidence shows we can consider more than one syntactic structure at a time (Cai et al., 2012).

✓ Wilson & Garnsey (2009) found that previous experience with verbs had an effect on sentence
processing.
✓ Brysbaert and Mitchell (1996) found there were individual differences among Dutch people in their
parsing decisions.

There are two main limitations to consider:

- It is difficult to make predictions about parsing, therefore difficult to falsify (Rayner et al., 2012).
- Research showing that non-syntactic structures happen early could also be supporting garden-path
model (Eysenck & Keane, 2020).
DISCOURSE PROCESSING - INFERENCES
Discourse processing involves processing extended segments of language
• We draw inferences to make sense of what we are listening to or reading
• There are three types of inferences:
1. Logical inferences (depend only on the meaning of words)
2. Bridging inferences (or backward inferences; conclusions drawn to increase clarity between current
and preceding parts of a text/speech)
3. Elaborative inferences (embellish or add details to the text, using world knowledge)
Anaphor resolution: a form of bridging inference, whereby pronoun or noun has to be identified with a
previous noun or noun phrase
- Example: Fred sold John his lawn mower, and then he sold him his garden hose
- Evidence shows gender info makes anaphor resolution easier (Arnold et al., 2000).
DISCOURSE PROCESSING – THEORETICAL APPROACHES
• There are several theories surrounding inferences, and here are some:
• Constructionist approach (Bransford et al., 1972)
- Readers usually construct a complete mental model of the situation and the events described in the text
• Minimalist hypothesis (McKoon & Ratcliffe, 1992)- Much fewer inferences are drawn
Search-after-meaning theory (Graesser et al., 1994)
- There is flexibility in the number of inferences drawn by readers.
• More support for minimalist and search-after-meaning theory, than constructionist approach (see
Eysenck and Keane, 2020, for a review).

UNIT 5: Eyewitness memory

WHAT IS EYEWITNESS MEMORY

The American Psychological Association (2020) defined eyewitness memory as “an individual’s
recollection of an event, often a crime or accident of some kind, that he or she personally saw or
experienced.”
• Eye-witness memory is of great importance – the judicial system often rely on eye-witness testimonies
when there is no DNA evidence. However, memory is not 100% reliable.

EYEWITNESS MISIDENTIFICATION

Memory is prone to errors and distortions, and even false memories.

• Most jurors and judges underestimate problems with eye-witness testimonies (Benton et al., 2006).

Albright (2007) - Eyewitness misidentifications can be characterised as:

1. Failures of visual perception
2. Failures of memory

CONFIRMATION BIAS

Confirmation bias is the tendency to process information by looking for, or interpreting, information that
is consistent with one’s existing beliefs (Casad, 2019)
• Examples (VeryWell Mind, 2021): Not seeking out objective facts and Interpreting information to
support your own beliefs
Remembering details upholding your beliefs and Ignoring information challenging your beliefs.
Lindholm & Christianson (1998) showed that participants were more likely to pick innocent immigrant
over innocent Swede to be the culprit in a robbery simulation

SCHEMAS
Bartlett (1932) proposed that people have schemata, or unconscious mental structures, which store
general knowledge of the world in long-term memory
• Schemas stored in long-term memory can lead us to form certain expectations – can distort memory by
causing us to reconstruct an event’s details based on ‘what must be true’ (Barlett, 1932)
• Tuckey and Brewer (2003) found that eye-witnesses recalled information from a simulated bank
robbery only if it matched their existing schema.
CHANGE BLINDNESS
Change blindness has been defined as an inability to notice changes in the environment (Simons &
Chabris, 1995; Simons & Levin,1998)
Research shows that people can miss large changes in a scene and this can result in misidentifying
innocent
individuals as perpetrators of crimes (Davies & Hine, 2007; Davies et al., 2008).
• Davies & Hine (2007) showed that if eyewitnesses are primed to pay attention to a crime scene, then
they are more likely to detect change and make a successful identification.
• Nelson et al. (2011) found that change blindness led to misidentification. Crime severity also affected
eyewitness accuracy (more severe crimes, led to fewer errors).

WEAPONS FOCUS
Weapons focus is the increased likelihood to attend to the weapon than other details of the event (Loftus
et al., 1987).
• Theoretical explanations of weapons focus:
- The arousal hypothesis (Loftus, 1980; Peters, 1988) which is supported by the Cue-Utilisation model
(Easterbook, 1959)
- Threat superiority effect (Fox, 2006; Eastwood, 2001)
- Unusual objects hypothesis (Shaw & Skolnick, 1994; Mitchell et al.,1998)
Loftus et al. (1987) showed that participants focused on a gun more than a cheque and remembered less
details unrelated to the gun in the gun scenario.
• Biggs et al. (2013) found that participants looked more at guns than faces. But if they were holding guns
themselves, in a readily usable position, they looked at faces more than objects.

STRESS AND ANXIETY

There is no doubt that this stress, anxiety or arousal influences the memory of an event
• The Cue-Ulitlisation model (Easterbrook Hypothesis - 1959): - stress causes narrowing of attention on
central or important information, but impairs memory for peripheral details (e.g. what the perpetrator
was wearing)
• Violence and trauma tend to improve memory for the central gist of an event (e.g. witnessing a murder),
but impair memory of the peripheral details of the event (e.g. clothing) (Loftus & Burns, 1982;
Christianson & Loftus, 1987)

Biological research:
- Stress & arousal can strengthen memories due to epinephrine & cortisol release (Roozendaal &
McGaugh, 2011)
- However, too much stress can result in reduced memory (Adreano & Cahill, 2006). Depends on
individual’s stress response
• Meta-analysis (Deffenbacher et al., 2004) found:
- Culprit face identification: 54% in low-anxiety conditions vs. 42% in high-anxiety conditions
- Average proportion of details remembered: 64% in low-stress conditions vs. 52% in high-stress
conditions
MISINFORMATION EFFECT
The misinformation effect is the distorting effect on eye-witness memory of misleading information
presented after a crime
• Terminology: - Proactive interference - old memories interfere with retrieval of new memories
- Retroactive interference - new memories interfere with the retrieval of old memories and information
The wording used by interviewers can affect the memory of witnesses (e.g. “smashed” instead of
“bumped”; Loftus & Palmer, 1974) (retroactive interference).
• Ecker et al. (2011) - The misinformation effect was still present when participants were warned not to
be influenced by misleading information
• Lindsay et al. (2004) - Misinformation before an event can also lead to distortion of memory (proactive
interference).
• Dalton & Daneman (2006) argued that misinformation effect only occurs with peripheral information,
not central information.

Theoretical explanations:
• Source misattribution (aka source monitoring errors): the misidentification of the origin of a memory
(Johnson et al., 1993). Evidence: Lindsay et al. (2004)
• Reconsolidation: misleading information reactivates the original memory, making it open to change.
Evidence: Chan & Lapglia (2013)
Vacant slot information: related information from the original event was not stored in memory (Wright &
Loftus, 2008).
• Blend explanation: information from the original event and misinformation are combined together in
memory (Wright & Loftus, 2008).
SOURCE MONITORING ERRORS (OR MISATTRIBUTION ERRORS)
• Inaccuracies in eyewitness memories can result from misattributing where the information came from
(Johnson et al., 1993; Johnson, 1997; Zaragoza & Lane, 1994)
• Three types of source monitoring:
1. External source monitoring – distinguishing between two different external information sources.
2. Reality source monitoring – distinguishing between what happened and what one imagined; also
discriminating between what was experienced (internal source) and what was communicated (external
source).
3. Internal source monitoring – distinguishing between one’s actions and thoughts
UNCONSCIOUS TRANSFERENCE
• This refers to when an eyewitness misidentifies a familiar but innocent person as the culprit (Loftus,
1976).
• Read et al. (1990) defined the concept as “the transfer of one person’s identity to that of another person
from a different setting, time, or context”
• Ross et al. (1994) reported unconscious transference can occur as a result of:
1. Automatic processing
2. Deliberate source-monitoring at retrieval
3. Conscious inference at encoding
SOCIAL FACTORS
• Interaction with other eyewitnesses can lead to misinformation effect (Gabbert et al., 2004)
• Eyewitnesses communicating together after a crime can (Gabbert et al., 2003; Paterson & Kemp, 2006;
Takarangi et al., 2006):
1. Reinforce common memories
2. However, it can also contaminate each other’s memories
VERBAL SHADOWING
• Putting an experience into words can result in failures of memory about that experience, whether it be
the memory of a person’s face, the colour of an object, or the speed that a car was going (Schooler &
EngstlerSchooler, 1990)
• Meissner & Brigham (2001) found that eliciting elaborate and detailed descriptions of a previously seen
face were more likely to demonstrate verbal overshadowing.

Theoretical explanations:
1. Recoding interference or content account (Schooler & Engstler-Schooler, 1990; Meissner et al., 2001):
generate and later rely upon an inadequate verbal description of the content of their original non-verbal
memory of the face.
3. Transfer-inappropriate processing shift account (Schooler et al., 1997; Schooler, 2002): verbalisation
encourages featural processing, rather than configural processing of the face, which is less effective (e.g.,
Valentine, 1988; Tanaka & Farah, 1993).
5. Criterion shift account (Clare & Lawandowsky, 2004): verbalization induces caution in people’s
responses.
OWN-RACE BIAS (OR SAME-RACE BIAS)
• People show better recognition memory for their own race compared to other races.
• Majority of eyewitness misidentifications may be due to own-race bias (Innocence Project, 2020)
• Contact Hypothesis: argues that we recognise own race better as we have had more exposure to faces
from the same race (Brigham & Malpass, 1985; Slone, Brigham & Meissner, 2000; Hugenberg, Miller, &
Claypool, 2007)
More exposure with other races reduces the own-race bias (Brigham et al., 2007).
• Wong et al. (2020) found that level of exposure to other-race faces only accounts for small part of own-
race bias.

AGE AND MEMORY

• Older eye-witness memory is less accurate than that of younger adults and misinformation effects are
much stronger in older adults (Jacoby et al., 2005).
• Own-age bias – more accurate identification when the culprit is similar age to themselves (Wright &
Stroud, 2002)
• Poss. due to greater exposure of people of their own age (Harrison & Hole, 2009).
• Weise et al. (2013) young geriatric nurses could recognises older faces better than young controls could.
CHILDREN’S EYEWITNESS TESTIMONY
What about children?
• Children more susceptible to suggestibility (Volpini et al., 2016; Kassin et al., 2001)
• Suggestibility in police interviews led to many eyewitness misidentifications in abuse cases (Garven et
al., 1998)
• Ceci and colleagues (1987) showed how young children are very suggestible to misleading information
Kulofsky and Klemfuss (2008) found that younger children more susceptible than older children to
misleading information
• Neutral interviewing associated with more accurate eyewitness memory (Thompson et al., 1997)
• Accuracy improves when children are asked to draw (Gross & Hayne,1999)

GENDER AND EYEWITNESS TESTIMONY

Females (Areh, 2011) -
1. Are more reliable witnesses
2. Outperform males in accuracy person description
• Males (Areh, 2011) -
1. Are more accurate in describing the event
• No gender differences were noted with regards to the quantity of details remembered (Areh, 2011)

UNFAMILIAR VS. FAMILIAR FACES

Ability to match unfamiliar faces much worse when photos taken months apart (Megreya et al., 2013).
• Jenkins et al. (2011) showed that photos from same face can be very different and when the person is
unfamiliar participants can mistakenly think they are different people.
• Burton (2011) argued that photos of same person should be combined to form an average face.

LINE-UP PRESENTATION

Simultaneous line-ups lead to more errors than sequential line-ups (Steblay et al., 2011).
• Misidentification is reduced when witnesses are given a ‘not-sure’ option (Steblay & Phillips, 2011).
• Telling the witness that the culprit might not be in line-up reduced mistaken identification (Steblay,
1997).
FROM THE LAB TO THE COURTROOM
Differences:
- Eyewitnesses likely to be victims in real life
- More stressful to be a witness to a violent crime
- Consequences of misidentification are much more serious in real life
• However:
- Lab conditions don’t distort findings (Ihlebaek et al., 2003)
- What we see in labs is an underestimate of what is happening in real life.

COGNITIVE INTERVIEWING - GEISELMAN & FISHER (1997)

• The APA (2020) defined the cognitive interview as “a structured technique developed for enhancing
eyewitness recollection in criminal investigation…”
• Based on 4 rules:
1. Eyewitness recreates the context of the crime.
2. Eyewitness reports everything he/she can remember.
3. Eyewitness reports details of incident in various orders.
4. Eyewitness reports events from various perspectives.

Evidence:
• Memon et al., (2010) conducted a meta-analysis comparing standard police interview with cognitive
interview, and found:
1. Large increase in number of details correctly recalled
2. Beneficial effect reduced in high arousing situation and long interval
3. Increased number of incorrect details

Vredeveldt et al. (2011) best done with eyes closed

• Memon et al. (2009) cognitive interview is not susceptible to misinformation after the interview, but is
before
• Dando et al. (2011) found that changing order was ineffective and made things worse.
Overall, to conclude:
• We cannot rely on eyewitness memory
• There are many factors that influence eyewitness memory
• There are strategies, such as cognitive interview that can aid minimising memory distortions in
eyewitness memory

UNIT 6: Face Perception

The APA (2020) defined face perception as “neurocognitive processes involved in the recognition of
individual faces and the interpretation of facial expression”
• Recognising faces is of great importance to us as a social species because:
- It helps us function in the world
- It enables us to connect with others
• Some species use olfaction or audition to recognise others, but primates do it via the visual system,
using faces
Face processing begins and birth (Pascalis et al., 2011), and our face processing abilities are highly
Specialised. Some argue it is an innate mechanism (e.g. Wilmer et al., 2010)
• Others believe it to be due to visual learning of faces through exposure to faces (e.g. Bukach et al., 2006)
• However, both genetic predisposition and exposure to faces appear to be important in face processing.

FACE PERCEPTION VS. OBJECT PERCEPTION

Face perception differs from other forms of object perception
• Face perception involves holistic processing:
- Processing that involves integrating information from a whole object and focusing on relationships
between features
- Faster and more efficient than feature processing

Evidence which supports the notion that faces are processed holistically include:
- Face inversion effect (McKone et al., 2006)
- Thatcher illusion (Thompson,1980)
- Part-whole effect (Tanaka & Farah, 1993)
- Composite-faces effect (Richler et al., 2011)
FACE INVERSION EFFECT
• Faces are harder to process when presented upside down
• McKone et al., (2006) found that same effect was not seen as strongly in objects
• Configural-processing hypothesis: inversion impairs the processing of configural information (the
spatial
relations between features), but doesn’t impair the processing of featural information (e.g., eyes, nose,
and
mouth) (Freire, Lee & Symons, 2000)

THATCHER ILLUSION EFFECT

• The Thatcher illusion effect was demonstrated by Thompson in 1980
• It becomes more difficult to detect local feature changes in an upside-down face, despite identical
changes being obvious in an upright face.

PART-WHOLE EFFECT
• The part-whole effect describes the difficulty people have in recognizing familiar faces from isolated
features, rather than when the features are within a face (Davidoff & Donnelly, 1990; Donnelly & Davidoff,
1999; Tanaka & Farah, 1993; Tanaka & Sengco, 1997)
• No part-whole effect found in objects (e.g. Tanaka & Farah, 1993).

COMPOSITE-FACE EFFECT
• Richler et al. (2011) showed that it is difficult to ignore bottom half of face
• Murphy, Gray & Cook (2016) composite-face effect diminishes when the two halves are misaligned
When aligned with different lower halves, it is surprisingly difficult to recognise that the upper regions of
the two composites are identical (left top). However, the illusion-induced interference is greatly
diminished when the composites are misaligned (right top). When composite arrangements are shown
upside-down,
little illusion induced interference is seen in either the aligned (left bottom) or misaligned (right bottom)
conditions
ARE FACES SPECIAL?
• The inversion, Thatcher, part–whole and composite effects provide evidence that faces undergo holistic
processing
• We have a lot of experience of faces more than any other object (for the majority of people)
• It is possible that holistic processing is found for any category of objects for which an individual
possesses expertise?
PROSOPAGNOSIA
• The APA (2020) defined prosopagnosia as "a form of visual agnosia in which the ability to perceive and
recognize faces is impaired, whereas the ability to recognize other objects may be relatively unaffected”
• The symptoms and severity of prosopagnosia can vary across individuals
• Often show signs of covert recognition (i.e. face processing without conscious awareness; Simon et al.,
2011).

Explanations include:
1) damage to specific face-processing brain regions.
2) Face recognition is harder, because it involves distinguishing between same category
• Busigny et al., (2010) tested both of these explanations (patient GG and controls)
• Moskovitch et al. (1997) found that patient (patient CK) with object agnosia performed as well as
controls
on face recognition tasks

FUSIFORM FACE AREA (FFA)

• An area of the brain associated with face processing in the inferior temporal cortex, within the fusiform
gyrus (Brodmann area 37)
• People with prosopagnosia often have damage to this area (Kanwisher & Yovel, 2006)
• Seems to be more active in processing faces, than other objects (Downing et al., 2006)

However there are some factors to consider…

• It is active for numerous other objects too, not just faces
• Face processing may involve a network of different regions, including fusiform face area and occipital
face
area (Gainotti & Mara, 2011; Atknson & Adolphs, 2011)

EXPERTISE HYPOTHESIS
• The FFA is probably involved in expertise, rather than face processing (Gautier and Tarr, 2002)
• Expertise leads to holistic (configural) processing.
• Predictions:
1) Holistic processing occurs for any object in which one has expertise
2) The FFA should be activated for any object category in which one has expertise
3) Young children will have less evidence of holistic processing of faces than older children and adults.
4) If the processing of faces and of objects of expertise involves similar processes, then objects of
expertise should interfere with face processing.

EVIDENCE FOR EXPERTISE HYPOTHESIS

Gauthier & Tarr (1997) created novel non-faces stimuli called ‘Greebles’ to train expertise. They found
that
Greeble expertise led to recognition effects typically associated with faces
FFA activation increased with Greeble expertise (Gauthier et al., 1999)
• However, Brants et al., (2011) reported that Greebles are too face like
Prediction 1: - Kundel et al., (2007): expert doctors typically focused immediately on the cancer in
mammograms, showing they use very fast holistic processing
- McKone et al., (2007) found expertise in other objects did not lead to holistic processing
Prediction 2: - McKone et al. (2007) found some tendency for fusiform face area to be more activated by
objects of expertise, but this was outside of fusiform area rather than inside it
- Bilalic et al. (2010) chess experts had greater activation in response to chess pieces, than novices, but
both had greater activation in faces, than chess
Prediction 3: - Wallis (2013) reported that the fusiform face area increases from childhood to early
adulthood.
- Crookes & McKone (2009) suggested that may be due to methodological problems
• Prediction 4: - McKeef et al., (2010) showed that car expertise interfered with face processing.

FACE PERCEPTION DEFICITS - NEURODEVELOPMENT

Autism Spectrum Disorder (ASD):
- Neurodevelopment condition
- Hypersensitivity (bright lights or loud noises feel overwhelming, stressful or uncomfortable) - NHS
(2018)
- OCD tendencies (NHS, 2018)
- Autism is a spectrum, and everybody with autism is different (NHS, 2018)
- Difficulties in communicating and interacting with others

Face processing deficits in Autism:

- One of the reasons for difficulties in communication and interaction might be the impaired face
perception that usually accompanies ASD (Bi & Fang, 2017)
- Perceptions of the face and their neural processing are significantly impaired in those with ASD (Nomi &
Uddin, 2015)
Neurophysiological evidence:
- Ewbank et al., (2017) reported that in the fusiform gyrus region, the effect of face stimuli was weaker in
an ASD group vs. control group
- The function of the left fusiform area has been found to be atypical in the face identity processing of ASD
individuals (Nickl-Jockschat et al., 2015)
• The face recognition task is a promising ASD diagnosis tool (Wang et al., 2015)
INDIVIDUAL DIFFERENCES
• With regards to facial scanning, Rennels & Cummings (2013) reported gender differences:
Males:
- Male infants were better at face recognition compared to female infants
Females:
- Female adults are better at face recognition than male adults
- Female infants and adults, discriminate emotional expressions better than males
MODEL OF FACE RECOGNITION - BRUCE & YOUNG (1986)
• Face recognition goes through various stages:
Predictions:
1. There should be major differences in the processing of familiar and unfamiliar faces
2. Separate processing routes are involved with facial expression and facial identity
3. You can find a face familiar without knowing anything else about them or you can know their
occupation without knowing their name

GAZE DIRECTION
• Gives strong cues about intentions and future actions (Baron-Cohen, 1995) by providing information
about where the person is attending to (Clifford & Palmer, 2018).
• Newborns look at faces whose gaze direction is looking towards them, rather than away (Farroni et al.,
2002).
• Infants smile more at direct gaze (Hains & Muir, 1996; Symons et al., 1998). Mutual gaze important for
learning of faces (Blss & Camp, 2001)

Theory of Mind (Baron-Cohen et al., 1985) suggested that interpretation of gaze plays important role in
normal functioning of theory of mind (Baron-Cohen, 1997)
• Research shows that gaze direction detection less accurate in autism spectrum disorder (Forgeot et al.,
2016).
WOLLASTON ILLUSION - WOLLASTON (1824)
• A person’s gaze is pulled in the direction of the head’s rotation
• Suggests that perception of gaze relies on integrating information from the eyes and rotation of head

FACIAL ATTRACTIVENESS
• Facial attractiveness refers to the pleasant emotional experience and approaching intention induced by
a human face (Rhodes, 2006)
• Highly attractive faces attract visual attention (Sui & Liu, 2009; Lindell & Lindell, 2014)
• Memory for highly attractive faces is superior to that of low attractive faces (Zhang et al., 2016), while
another study found opposite results (Wiese et al., 2014)

Attractiveness halo’ or ‘beauty is good effect’: people with more attractive faces are judged more
positively on a host of dimensions (Berscheid & Walster, 1972; Dion et al., 1972; Eagly et al., 1991;
Feingold, 1992)
• Attractive people are given preferential treatment in many life domains (Langlois et al., 2000)
WHAT FACES ARE CONSIDERED ATTRACTIVE?
• Certain facial features are considered more attractive than others:
- Symmetry (Rhodes et al., 1998)
- Averageness (Langlois & Roggman, 1990; Grammer & Thornhill, 1994)
- Distinct faces (Perrett, May, & Yoshikawa, 1994)
- Happy faces (Otta et al., 1996; Ebner et al., 2018)
Certain facial features are considered more attractive than others:
- Sexually dimorphic cues (Perrett et al., 1998; Cunningham et al., 1990).
- Youthfulness (Mathes et al., 1985; Alley, 1988; Henss, 1991, Zebrowitz et al., 1993)
• Strong evidence exists to show that raters agree within and across cultures who is and who is not
attractive (Langlois et al., 2000)

FACIAL ATTRACTIVENESS - AN EVOLUTIONARY PERSPECTIVE

• Evolutionary explanations believe attractive faces indicate good genes and good health (see review by
Little, Jones, DeBruine, 2011)
- Clear eyes and smooth skin relate to good health (Tovee et al., 1999)
- Attractive women indicate high fertility (Hill & Hurtando, 1996)
- Symmetrical faces are a signal of mate quality (Rhodes et al., 1998; Jones et al., 2001)

Average faces denote genetic variation, which protect against disease (Lie et al., 2008)
- Sexually dimorphic cues indicate fertility (Perret et al., 1998) and good genes (Perret et al., 1998;
Penton-Voak, Jacobson & Trivers, 2004)
• However, there are some criticisms to consider, Some criticisms relating to the evolutionary perspective
of facial attractiveness includes:
- Good looks don’t always equal good health or fertility (Kalick et al., 1998)
- Many symmetrical, youthful, and sexually dimorphic faces are not attractive (Langlois et al., 1994;
Rubenstein et al., 2002).

FACIAL ATTRACTIVENESS - A COGNITIVE PERSPECTIVE

• What is the cognitive perspective as to why we find certain facial attributes more attractive?
• Cognitive averaging theory (Langlois & Roggman, 1990)
- Attractive faces are those which are closer to facial prototypes
- Prototypical face: a face which closely resembles the majority of other faces in the population
- Prototypical faces are familiar to us
Preference for average faces is shared across cultures (Apicella et al., 2007)
- Prototypical faces are processed more easily, resulting in increased positive affect in the viewer (Shein
et al., 2017)
- Highly attractive and mathematically averaged faces were processed more fluently than unattractive
and nonaveraged faces (Trujillo, Jankowitsch & Langlois, 2014)

The perceptual bias account (Enquist & Arak 1994; Enquist & Johnstone 1997; Enquist & Ghirlanda,
1998)
- Symmetry is found attractive as a by-product of the relative ease with which the perceptual system can
process all symmetrical stimuli
- May be a result of the symmetrical nature of the human visual system (Mach, 1897; Herbert &
Humohrey,
1996)
- We also prefer symmetrical objects (Rentch, 1963) and decorative art (Gombrich, 1984)
Prototypical faces are symmetrical (Jones et al., 2003)
- Predictions: 1) should prefer symmetry regardless of orientation (i.e. upright or inverted); 2) should
prefer
familiar faces over symmetrical ones.
- Jones, Little & Perret (2003) found symmetrical upright faces are more attractive, than inverted
symmetrical faces. Also, symmetrical faces preferred over familiar faces.

UNIT 7: Attention
Attention normally refers to selectivity of processing (James, 1890)
• “Attention is…taking possession of the mind, in clear and vivid form, of one out of what seem several
simultaneously possible objects or trains of thought. Focalisation, concentration, of consciousness are of
its essence” (James, 1890, pp.403-404).
• The American Psychological Association (2020) defined attention as “a state in which cognitive
resources are focused on certain aspects of the environment…"

Attention can be
1. Active (endogenous or top-down processing)
- E.g. When you are looking for your keys; Paying attention to this lecture
2. Passive (exogeneous or bottom-up processing)
- It is controlled by external stimuli
DISTINCTIONS AND DEFINITIONS
Focused attention or selective attention:
- Focusing on one stimulus, in the face of other distracting stimuli
• Divided attention (e.g. dual tasking or multitasking):
- Paying attention to more than one thing at the same time
External attention:
- Paying attention to things in the environment
• Internal attention:
- Paying attention to internally generated information

AUDITORY AND VISUAL ATTENTION

We can attend to stimuli via any sense, but research is mostly carried out via our sense of audition
and vision
THE COCKTAIL PARTY PROBLEM - CHERRY (1953)
How can we follow one conversation when several people are talking at once?
• Listeners face two problems (McDermott, 2009)
1. Sound segregation
2. Sound source of interest
• How much processing is there of unattended input?
FOCUSED AUDITORY ATTENTION METHODOLOGY
Dichotic listening task:
- A different auditory message is presented to each ear and attention has to be directed to one message
FOCUSED AUDITORY ATTENTION METHODOLOGY
• Shadowing:
- Participant repeats one auditory message word for word, as it is presented, while a second auditory
message is also presented

THE COCKTAIL PARTY PROBLEM (CHERRY, 1953) - FINDINGS

Cherry (1953):
- Listeners solved the problem by making use of differences between auditory inputs in physical
characteristics
- Listeners found it difficult separating on the basis of content/meaning
- Very little info extracted from unattended message
- Physical changes always noticed
• Moray (1959) - Very little memory for unattended words.
BOTTLENECK THEORIES
Broadbent’s Filter Theory (1958)
- Two or more stimuli presented at the same time gain access to a sensory buffer simultaneously
- One of the inputs is allowed through a filter on the basis of its physical characteristics
The other input remains in the buffer for later processing
- Unattended info filtered out before the meaning is processed
Prediction: that there should be little or no processing of unattended auditory messages
Treisman’s Attenuation Theory (1964)
- Bottleneck location is flexible and the filter attenuates the analysis of unattended information
- Listeners start with processing based on physical cues, syllable pattern and specific words
- Then they move onto processing grammatical structure and meaning
- If processing capacity is insufficient to allow full stimulus analysis, then later processes are omitted
- Top-down processes also important
In shadowing task, breakthroughs mostly occur when the word in unattended input is highly probable in
the context of the attended message
• Prediction: the amount of processing of unattended auditory messages is flexible.
Deutsch & Deutsch (1963)
- All stimuli are fully analysed, with the most important or relevant stimulus determining the response
- Places the bottleneck nearer the response end – favours late selection of stimuli
• Prediction: there should be reasonable thorough processing of unattended stimuli.

EVALUATION OF BOTTLENECK THEORIES

• Treisman & Riley (1969) found that many more target words were detected on the shadowed message
• Bentin et al. (1995) found unattended words received some semantic processing, but the extent was
less than for attended words
• Processing of unattended words is enhanced if they have personal significance to the listener (Li et al.,
2011)
More likely to detect our own name when presented in unattended stimuli (Moray, 1959) – but this also
depends on our WM capacity (Conway et al., 2001)
• EEG data shows greater neural activity for target words when they are presented in the attended
message than in the unattended message (Coch et al., 2005).
• Horton et al. (2013) found that neural activity showed greater enhancement of the attended message
combined with suppression of the unattended message
Bottom-up cues, such as temporal coherence, play a role (Shamma et al., 2011)
• Top-down processes also play a role:
- Easier to identify target voice when words form sentence, than when they are random (McDermott,
2009)
- Familiarity with target voice (McDermott, 2009)
• Visual information also helps (Golumbic et al., 2013)
To conclude:
• More support for Treisman’s theory and least for Deutsch and Deutsch’s theory
• Early research on focused auditory attention using the shadowing task suggested very limited
processing of unattended stimuli. However, unattended stimuli also receives some processing
• Semantic processing can occur of unattended stimuli, therefore Broadbent’s theory not supported.
Factors influencing the processing of unattended stimuli:
- When unattended stimuli is dissimilar to the attended stimuli. When the meaning of the message is
salient
• Limitations- Performance in dichotic listening tasks rely on experience of listener (Eysenck & Keane,
1990; Moray, 1959)
- All three theories difficult to test, because difficult to know if participants have shifted attention to
unattended stimuli.

VISUAL SEARCH
Overt attention: Orienting attention via the movement of eyes towards the focus
- Measured by: eye-tracker (saccades and fixation points)
• Covert attention: Orienting attention to a different part of the visual field without moving the eyes or
body to do so
- Measured by: filtering tasks.
Feature Integration Theory (Treisman & Gelade, 1980; Treisman, 1998):
• There are two processing stages during visual search
1. Basic visual features (e.g. shape and colour) are processed rapidly and preattentively, in parallel across
the visual scene
2. Slower, serial process with focused attention providing ‘glue’ to form objects from features.

Illusory conjunction: mistakenly combining features from two different stimuli to perceive an object that
is not present (Eysenck & Keane, 2012)
• Prediction: Targets defined by single feature (e.g. a green letter or a T) should be detected rapidly,
whilst
conjunction of features are slower to detect and require attention.
Treisman & Gelade (1980) and Treisman (1998) visual search findings:
1. Items distinguished by a core feature (feature condition) pop out regardless of distractors
(bottom-up processing)
2. Conjunctive condition is more effortful as targets share features with distractors (top-down
processing)
3. As distractors increase in conjunction search, the reaction time increases, but no change in
feature search
• Lack of focused attention can produce illusory conjunctions (Friedman-Hill et al., 1995)

LIMITATIONS OF THE FEATURE INTEGRATION THEORY

Features defined retrospectively based on eliciting pop-out in visual search tasks
• Pop out may not be preattentive, it may reflect a stimulus driven cue of attention (Wolfe, 2003)
• Visual search task may be led by the distractors rather than the target (Duncan & Humphreys, 1989)

DIVIDED ATTENTION: DUAL-TASKING AND MULTITASKING

High multitaskers are easily distracted (Ophir et al., 2009).
• High multitaskers develop breadth-based cognitive control (not selective in what they pay attention to
(Ophir et al., 2009)
Low multitaskers were able to make use of top-down instructions to reduce distraction and enhance
performance (Cain & Mitroff, 2011).
• High multitaskers are more efficient task switchers (Alzahabi & Becker, 2013).
Factors that determine dual/multi-task performance include:
• Task similarity: - If the two tasks that are being performed are too similar or are using the same
modality (e.g. visual) then performance is hindered (Treisman and Davies, 1973)
- Similar in response modality McLeod, 1977)
DIVIDED ATTENTION: SERIAL OR PARALLEL PROCESSING?
• Serial processing: - Involves switching attention back and forth between two tasks with only one task
attended to and processed at a time
• Parallel processing: - Involves attending and processing both tasks at same time
• Processing is flexible (Lehle et al., 2009; Lehle & Hubner (2009)

DIVIDED ATTENTION: AUTOMATIC PROCESSING

Divided attention tasks improve with practice. This might be due to the tasks becoming automatic.
• Shiffrin & Schneider (1977):
- Controlled processes are of limited capacity, require attention and can be used flexibly
- Automatic processes have no capacity limitation, do not require attention and are difficult to change
once learned
- Automatic processes function in parallel, but are inflexible; controlled processes are flexible, but
function in slow, serial manner
• However, Shiffrin & Schneider’s (1977) too oversimplified; preferable to think in terms of degrees of
automaticity (Eysenck & Keane, 2012)

PSYCHOLOGICAL REFRACTORY PERIOD (PRP) EFFECT

PRP designed to test dual-task interference
• PRP effect:
- There are two stimuli (e.g. two lights) and two responses (e.g. button presses)
- The task is to respond to each stimulus as quickly as possible
- When the second stimulus is presented very shortly after the first stimulus, there is a significant slowing
of the response to the second stimulus
PRP effect occurs after 1000 trials (Pashler, 1993)
• Explanation: Task performance involves 3 sequential stages (perceptual, central and motor). Bottleneck
prevents more than one central decision process (Welford, 1952)
• Hibberd et al. (2013) found in a driving simulation task, drivers braking time was slowed in dual-task
conditions (they were distracted right before a car in front put the brakes on).

LIMITATION OF ATTENTION: ATTENTIONAL BLINK

A task used to demonstrate limitations of attention
• It shows that we have limited ability to report multiple targets that are distributed in time
• It is measured using Rapid Serial Visual Presentation tasks (RSVP), where participants often fail to
detect
a second target occurring in succession if it is presented between 200-500ms after the first one.
1. Filter-based accounts: filter out non-relevant visual information (e.g., Di Lollo, Kawahara, Ghorashi, &
Enns, 2005; Raymond et al., 1992)
2. Consolidation bottlenecks: a bottleneck in short term memory cache (e.g., Chun & Potter, 1995; Sergent,
Baillet, & Dehaene, 2005)
3. Retrieval complications: errors in the retrieval of information from a short-term memory cache (e.g.,
Shapiro et al., 1994)

ATTENTION RELATED IMPAIRMENTS

Spatial neglect
- Neglect is mostly a disorder of attention (Li & Malhotra, 2015)
- Patients fail to orientate, report or respond to stimuli located on the contra-lesional side (Li & Malholtra,
2015)
- It can be caused by many factors
- Commonly associated with stroke (Stone et al., 1993)
Neglect is more common and longer lasting after right hemisphere stroke (due to the role of the right
hemisphere in attention processing) (Mesulam, 1981; Li & Malhotra, 2015)
• Most patients recover, but they continue to have major cognitive deficits, particularly associated with
attention
• There are treatments that may improve neglect, however, a universal approach to therapy is yet to be
established (Li & Malhotra, 2015)

UNIT 8: Semantic Memory, Knowledge Representation and Concepts

Knowledge is represented in semantic memory
• Semantic memory is structured around concepts
• Knowledge representation is studied through exploring organisation of concepts in memory
• Lets explore this in a bit more detail

DEFINITIONS - SEMANTIC MEMORY

A form of long-term memory consisting of general knowledge about the world
• Memory for general factual knowledge and concepts (APA, 2020)
• Allows people to engage in complex cognitive processes (APA, 2020)

SEMANTIC MEMORY INDIVIDUAL DIFFERENCES

Gender:
- No differences between men and women on tasks assessing semantic memory (Herlitz, Nilsson &
Backman, 1997)
- Males had high means in general knowledge compared to females (Lynn & Irwing, 2002)
- A female advantage has been noted with regards to semantic memory (Maitland et al., 2004)
Culture:
- The contents of semantic memory do differ across cultures (Yoon et al., 2004)
- Cultural ideas and practices guide what semantic information is learned (Yoon et al., 2004)
- Cultural context’s influence how semantic information is employed and organised (Ji et al., 2004;
Unsworth et al., 2005; Gutchess et al., 2010)

DEFINITIONS - CONCEPTS
Mental representations of categories (e.g. of objects or items) stored in semantic memory
• Basic units of cognition
• Combine sensory and motor experiences with the environment into categories (Humphreys et al.,
1988; Kiefer & Pulvermuller, 2012)
• Includes concrete objects and referents (e.g. abstract ideas) (Harpaintner et al., 2018)

HIERARCHIES OF CONCEPTS
• Rosch et al. (1976): Concepts are organised hierarchically:
We tend to use basic-level categories the most:
- Right balance of informativeness and distinctiveness (Rosch et al., 1976)
- Most general level at which we use similar motor movements to interact with category members
(Bourdais & Pecheux (2009)
- Children learn nouns at this level first (Bourdais & Pecheux, 2009)

METHODOLOGY
Category verification task - Participant has to respond to whether a word belongs to a certain category or
not
• Feature verification task - Participant decides if features belong to a word presented to them or not
• Sentence verification task - Participant is asked to make quick judgments about sentences

WHEN BASIC LEVELS ARE NOT PREFERRED…

Expertise:
- Tanaka & Taylor (1991) found that experts use subordinate names for their area of expertise
- Anaki & Bentin (2009) found matching of faces occurred faster at the subordinate level, because we are
experts at processing faces

IS BASIC LEVEL ALWAYS FASTEST?

Familiarity:
- Anaki & Bentin (2009) found that familiarity with famous towers (e.g. Eiffel Tower) leads to fast
subordinate categorisation, as opposed to unfamiliar towers, which lead to basic-level categorisation
Press et al., (2013):
- Participants had to categorise photos of objects very quickly
- Categorisation was at superordinate (animal or vehicle), basic (cat or dog), or subordinate level
(Siamese cat or Persian cat)
- Performance most accurate and fastest at superordinate level
- Least accurate and slowest at subordinate level
- Categorisation at superordinate level faster than basic, because basic contains more information (Close
& Pathos, 2012).

HIERARCHICAL NETWORK MODEL (COLLINS & QUILLIAN, 1969)

Cognitive economy: Information stored at one level not repeated at lower levels. Assumed that lower
nodes inherit higher characteristics (property inheritance)
Evidence for: - Category size effect
• Evidence against: - Typicality effect (Smith et al., 1974); Familiarity effect

SPREADING ACTIVATION MODEL (COLLINS & LOFTUS, 1975)

Assumes that semantic memory is organised by semantic relatedness, rather than by a logical hierarchy
• Whenever a person thinks of concept, that node is activated
Activation spreads to related concepts
• Spreading activation decreases the further away it gets from original point of activation
Strengths:
- Explains typicality effect
- Explains priming effect (Meyer & Schvaneveldt, 1976)
• Limitations: - Flexibility reduces specificity so difficult to test

FEATURE COMPARISON MODEL (SMITH, SHOBEN & RIPS, 1974)

In the feature-comparison model by Smith, Shoben and Rips (1974) the emphasis is on semantic features
or properties of concepts
• There are two types of features:
1. Defining features (essential features, e.g. ‘red breast’ of a robin)
2. Characteristic features (associated, but not necessary to define, e.g. ‘flies’ of a robin)

When comparing two concepts (e.g. is a robin a bird?) the following decision-making process is used. All
features are compared to determine similarity
- If highly similar – quick decision - true
- If highly dissimilar – quick decision - false
- If medium similarity - compare defining features – true/false

THE PROTOTYPE VIEW

Prototype: A set of characteristic attributes or best example of a concept (e.g. Hampton, 1979; Psner &
Keele, 1968; Rosch, 1978)
• Categories contain fuzzy boundaries
• Instances within a concept can be ranked according to typicality
• Category membership determined by the similarity between object’s attributes and prototype of that
Category
Evidence for:
- Colour categories are universal (e.g. Berlin & Kay, 1969)
- Typicality effect (e.g. Smith & Medin, 1981)
- Basic-level organised around prototype.
Evidence against:
- Not all concepts have prototypic characteristics (Hampton, 1981)
- Why some attributes are important and others aren’t? (Malt & Smith, 1983)
- Some members of category are not similar (Murphy & Medin, 1985).

SITUATED SIMULATION THEORY (BARSALOU, 2009, 2012)

• Barsalou (2009, 2012) argued traditional assumptions of concepts are incorrect and that we rarely
process concepts in isolation
• Situated simulation theory proposes that perceptual and motor systems are normally used in concept
Processing

SITUATED SIMULATION THEORY (BARSALOU, 2009, 2012) - EVIDENCE

• Wu & Barsalou (2009) found that conceptual processing can involve the perceptual system (the senses)
when participants are asked to name properties of nouns
• Wang et al. (2010) carried out a meta-analysis of neuroimaging studies and found that concrete objects
processing were associated with brain regions involved in perceptual processing, whilst abstract
concepts were not
Wilson-Mendenhall et al. (2013) instructed participants to think deeply about abstract concepts. Further
brain regions were found to be involved, depending on the concept.
Barsolou & Wiemer-Hastings (2005) - much knowledge of abstract concepts is quite concrete and
involves perceptual properties
• Hauk et al. (2004) found that verbs activated parts of the motor strip involved with movements
associated with the verbs (e.g. ‘lick’ activated part of motor strip involved with tongue movements)
• Processing of abstract concepts involves the motor system (e.g. Glenberg & Kaschak, 2002)

SITUATED SIMULATION THEORY (BARSALOU, 2009, 2012) – EVALUATION

Strengths:
- Much support for the theory
- Embodiment or grounded cognition: “an effect where the body, its sensorimotor state, its morphology,
or its mental representation play an instrumental role in information processing” (Korner et al., 2015, p.
1)
- Explains why concept processing varies across situations depending on the goal
Limitations:
- Exaggerates how much concept processing varies in different concepts (Mazzone & Lalumera, 2010)
- Both theoretical approaches are partially correct (Eysenck & Keane, 2015)
- The finding that concept processing involves both perceptual and/or sensory features may have other
interpretations (Eysenck & Keane, 2015). BUT, much evidence suggesting these processes are needed for
concept meaning (Pulvermuller, 2013)

HUB-AND-SPOKE MODEL
Patterson et al. (2007), Patterson (2016) and Probric et al. (2010) proposed hub-and spoke models of
semantic representation
• Spokes consist of modality-specific regions involving sensory and motor processing
• Hubs unify and integrate knowledge and are modality-independent
• Hubs are located in anterior temporal lobes (Patterson et al., 2007)
Evidence for hubs:
- Binder et al. (2009) found that ‘hubs’ are located in anterior temporal lobes and related regions
- Findings from semantic dementia patients (who have anterior temporal lobe deterioration) show that
they struggle with categorising atypical and pseudotypical items, suggesting hubs are more general
(Mayberry et al., 2011)
Evidence for spokes:
- Category-specific deficits: disorders caused by brain damage whereby semantic memory is disrupted for
certain semantic categories (Eysenck & Keane, 2015)
- Patients with category-specific deficits struggle with living things over non-living things and this may be
due to the different properties of living and non-living things (Marques et al., 2013).
- Cree & McRae (2003) identified 7 patterns of category specific deficits
- Pobric et al. (2010) found slowing in naming times to objects that could be manipulated when TMS
applied to inferior parietal lobule (involved in actions towards objects) to inhibit its function.
Limitations:
- Left anterior temporal lobe involved in language processing, so might not be as general as is assumed
(Mesulam et al., 2013)
- Unclear exactly what info is in hubs (Eysenck & Keane, 2015)
- Unclear how spokes integrate with hubs (Eysenck & Keane, 2015)
- Number and nature of spokes unknown (Eysenck & Keane, 2015)

LEARNING NEW CONCEPTS?

Learning in one context does not transfer (generalise) to other context’s, unless the situations are similar.
• Barnet and Ceci (2002) reported that transfer is less likely if the context differs in any of these ways:
- Knowledge domain
- Physical context (environment)
- Temporal context (time)
- Functional context (purpose)
- Social context
- Modality
Factors promoting generalisation include:
- Slower and deeper learning (Schmidt & Bjork, 1992)
- A hybrid approach (to balance efficiency and transfer)

SCHEMAS
In psychology and cognitive science, a schema describes a pattern of thought or behavior that organizes
categories of information and the relationships among them
SCHEMAS IMPACT MEMORY RETRIEVAL
• Bartlett (1932) argued that schemas play a crucial role in how we remember stories or events
• Brewer and Treyens (1981) found that participants remembered more schema-consistent objects than
schema inconsistent objects when participants were asked to remember objects in a ‘graduate student’s
room’
• Steyvers & Hemmer (2012) found that schema-inconsistent objects were recalled well too
• Van Kesteren et al. (2012) found that schema-inconsistent information is processed in the medial
temporal lobe
ASSUMPTIONS OF SEMANTIC MEMORY
• Semantic memory contains two types of information:
1. Abstract concepts generally correspond to individual words
2. Broader and more flexible organisation structures based on schemas
DISTINCT TYPES OF SEMANTIC MEMORY?
• Bier et al. (2013) tested script memory in patients with semantic dementia. Findings: 2 struggled, 1
didn’t
• Sirigu et al. (1995) found patients with prefrontal damage couldn’t put event related actions of script in
order. Prefrontal cortex involved in goal-directed behaviour.
Consentino et al. (2006) found that patients with fronto-temporal dementia struggled with sequencing
errors, but not semantic knowledge of concepts
Conclusion:
- Research supports notion that there is a distinction between concept and script knowledge
- However, the two are related, thus semantic dementia patients can have problems with accessing script
knowledge.

UNIT 9: Heuristics & Biases in Judgement and Decision Making

Judgement: Deciding on the likelihood of various events using incomplete information.
Decision making: Selecting one option from several possibilities. If information is unavailable then
judgement is necessary.
Heuristics: “Strategies that ignore part of the information, with the goal of making decisions more quickly,
frugally, and/or accurately than more complex methods” (Gigerenzer & Gaissmaier, 2011, p. 454).
Bias: A disproportionate weight in favour of or against something.

REPRESENTATIVE HEURISTICS
• The assumption that an object or individual belongs to a specified category because it is typical of that
category (Tversky & Kahneman, 1974)
• We ignore base-rate information (the relative frequency with which an event occurs within a
population)
• Conjunction fallacy: The mistaken belief that the conjunction or combination of two events is more
likely
than one event on its own (Tversky & Kahneman, 1983)
- Linda is 31 years old, single, outspoken and very bright. She majored in philosophy. As a student, she
was deeply concerned with issues of discrimination and social justice, and also participated in anti-
nuclear demonstrations.
• Tentori el al., (2013) argued that the conjunction fallacy happens because the hypothesis is supported
by the description

AVAILABILITY HEURISTICS
The frequencies of events can be estimated accurately by the subjective ease with which they can be
retrieved (Tversky and Kahneman, 1974)
• Litchenstein et al., (1978) asked participants to judge the relative likelihood of different causes of death.
Participants judged murder to be more likely than suicide, because murder attracts more public attention
(the opposite is true)
Pachur el al., (2012) three possible explanations for the judgement on how likely each cause of death is:
1. Availability heuristic based on own experiences
2. Availability heuristic based on own experience and media coverage
3. Affect heuristic (emotion influencing judgement)
• Availability heuristic can be over-ridden using deliberate thought (Oppenheimer, 2004), but not when
under cognitive load (Oppenheimer & Monin, 2009).

ANCHORING AND ADJUSTMENT HEURISTIC

Taking an initial number as a starting point (anchor) and then adjusting it upward of downward from
this starting point (Tversky and Kahneman, 1974)
• If the value is varied then the estimate changes (Tversky & Kahneman, 1974)
1. Participants were asked to estimate the percentage of African countries in the UN
2. They were given a number between 0-100, determined by a spinning wheel
3. Participants who were given 10, estimated 25%
4. Participants who were given 45, estimated 65%.
Pricing decisions (Northcraft & Neale, 1987), estimates for prime interest rates (Russo & Schoemaker,
1989) and negotiation (Liebert el al., 1968) are all susceptible to anchoring
• Selectivity accessibility model (Mussweiler & Strack, 1999) assumes that we compare the target with
the anchor by testing the possibility that the target’s value is equal to the anchor value

FAST-AND-FRUGAL HEURISTICS
• “If we are so dumb, how come we’re so smart?” (Glymour, 2001, p. 8)
• Heuristics can often be very valuable (Gigerenzer & Gaissmaier, 2011)
• Take-the-best heuristic: ‘take-the-best, ignore the rest’
FAST-AND-FRUGAL HEURISTICS
With regards to ‘take-the-best heuristic: ‘take-the-best, ignore the rest’:
• Three components:
1. Search rule
2. Stopping rule
3. Decision rule

FAST-AND-FRUGAL HEURISTICS: RECOGNITION

HEURISTIC
• ’When presented with two options we prefer to go with one we recognise the most
• Goldstein and Girgerenzer (2002) argued that when an object is recognised, but the other one isn’t no
other information influences the judgement
• American students were shown pairs of German cities and they had to decide which was larger. When
one city was recognised they used recognition heuristics (or did they?) (Goldstein & Gigerenzer, 2002)
• Not useful for judging which of two cities is further north (Eysenck & Keane, 2020)
• Richter and Spath (2006) found that recognition heuristic is not always used when they had access to
inconsistent information
• Newell (2011) argued that people also consider why they recognise an object and then decide if they
should use recognition heuristic
FAST-AND-FRUGAL HEURISTICS: HIATUS HEURISTIC
• Rule of thumb that only customers who have purchased goods recently remain active customers.
• Wubben and van Wangenheim (2008) found that hiatus heuristic that uses limited information
correctly categorised 83% of customers, and a complex model that used available information categorised
75% of customers
• Less information was more effective

DUAL PROCESS THEORY - KAHNEMAN (2003)

Rule of thumb that only customers who have purchased goods recently remain active customers
1. System 1: “…typically fast, automatic, effortless, implicit (not open to introspection) and emotionally
charged; they are also difficult to control or modify.”
2. System 2: “…slower, serial (one at a time), effortful, more likely to be consciously monitored and
deliberately controlled; they are relatively flexible and potentially rule governed.”
EVALUATION OF THE DUAL PROCESS THEORY
De Neys (2006) found that participants who were accurate took longer which suggests they were using
System 2, when making a decision about the Linda problem
• De Neys (2006) found that carrying out a dual-task whilst answering Linda-like problems led to poorer
performance, supporting the notion that system 2 is cognitively effortful
• Le Mens & Denrell (2011) showed that people can make errors when using system 2

• The theory assumes that system 2 is needed for answers that reflect base-rate information, but this is
incorrect (Pennycook & Thompson, 2012)
• The theory has been criticised for being to neat and tidy, with no real evidence that the two systems are
distinct (Keren & Schul, 2009; Fielder and von Sydow, 2015)

PROSPECT THEORY (KAHNEMAN & TVERSKY, 1979, 1984)

Proposed to explain paradox of why people avoid gains when the odds are in their favour
It is assumed:
1. Individuals identify a reference point representing their current state
2. Individuals are more sensitive to potential losses than to potential gains; this is loss aversion
• People overweight low probability events and underweight high probability events
FRAMING EFFECT
• Irrelevant aspects of a situation can influence our decisions (e.g. wording)
• Tversky and Kahneman (1981) used the Asian disease problem to study this:
Imagine that the US is preparing for the outbreak of an unusual Asian disease, which is expected to kill
600 people. Two alternative programmes to combat the disease have been proposed. Assume that the
exact scientific estimate of the consequences of the programme are as follows
In gain-frame condition, participants chose between:
- If programme A is adopted, 200 people will be saved
- If programme B is adopted, there is a 1 in 3 probability 600 people will be saved and a 2 in 3 probability
that no people will be saved
• In loss-frame condition, participants chose between:
- If programme C is adopted, 400 people will die
- If programme D is adopted, there is a 1 in 3 probability that nobody will die, and 2 in 3 probability that
600 people will die. If programme A is adopted, 200 people will be saved
Results of Asian problem study (Tversky and Kahneman, 1981):
- 72% of participants chose the certain gain (programme A) even though the two programmes (if
implemented several times) would lead to saving 200 lives on average - 78% of participants chose
programme D, even though the two programmes would have the same effect if implemented several
times
This means how something is framed or worded can affect the decision
• When programmes were described in full framing effect disappeared (Mandel and Vartanian, 2011)
• Wang (1996) argued that social and moral factors also play a role
• Almashat et al., (2008) found framing effect disappeared when people weigh up pros and cons and
justify decision
SUNK-COST EFFECT
• “A tendency for people to pursue a course of action even after it has proved suboptimal, because
resources have been invested in that course of action” (Braverman & Blumenthal-Barby, 2012, p. 186)
• Dawes (1998) illustrates the sunk-cost effect by presenting participants with a scenario about a couple
who fall ill before going on holiday and have paid $100 non-refundable deposit – what should they do?

Balgia and Ely (2011) argue that participants who remember all information should be immune from this
effect
• Braverman & Blumenthal-Barby (2012) showed that expert healthcare providers who had to make
treatment recommendations did not exhibit a sunk-cost effect

OVERWEIGHTING RARE EVENTS

• Hertwig et al., (2004) compared decision making based on descriptions or based on personal
experience. It was found that participants overweighted probability of rare events when decisions were
based on descriptions, but underweighted based on experience
• Hillbig and Glockner (2011) found that participants did not overweight small probabilities when the
format in which information was presented to them was rapid and involved a large amount of detail (they
called this open sampling or experience-based).
Conclusion:
- Weighting of rare events in decision making relies on precise information provided
- Findings from sampling and sequential sampling are inconsistent with Prospect Theory.
EVALUATION - PROSPECT THEORY
Better account of decision making than utility theory
✓ Explains framing, sunk-cost and loss aversion
No explicit rationale for the value function
- Over-simplified theory
- Overweighting of rare events doesn’t always occur.
- Individual differences play a role (Josephs et al., 1992)
- Doesn’t take social and emotional factors into consideration (Wang, 1996)
EMOTIONAL BIASES
• Impact bias: overestimation of the intensity and duration of negative emotional reactions to loss
(Kermer et al., 2006)
• Schlosser et al., (2013) distinguished between two types of emotions:
1. Anticipated emotions
2. Immediate emotions

Omission bias: a preference for risking harm through inaction compared to risking harm through action
- Brown et al., (2010) and Wroe et al., (2005) support this in studies of parents making decisions about
getting children vaccinated
• Status quo bias: a preference for maintaining status quo (present state) rather than acting to change
one’s decision
- Samuel and Zeckhauser (1988) supports this bias.
SOCIAL BIASES
• Tetlock (2002) proposed the social functionalist approach: people behave like intuitive politicians, as in
they need to justify their decisions to others
• Decisions with high accountability were more likely to show high sunk-cost effect (Simonson and Staw,
1992)
• Medical experts decision making is more biased when forced to be held accountable (Schwartz et al.,
2004)
• Limitations:
1. Individual differences not taken into account
2. Difficult to relate to real world