See

discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/231876489

The Visual Dynamics of Upper Palaeolithic Cave

Art

Article in Cambridge Archaeological Journal October 2008

DOI: 10.1017/S0959774308000401

CITATIONS READS

12 92

1 author:

Derek Hodgson
The University of York
57 PUBLICATIONS 328 CITATIONS

SEE PROFILE

All content following this page was uploaded by Derek Hodgson on 30 April 2015.

The user has requested enhancement of the downloaded file. All in-text references underlined in blue are added to the original document
and are linked to publications on ResearchGate, letting you access and read them immediately.
 The Visual Dynamics of Upper Palaeolithic Cave Art.

By
Derek Hodgson

Pre-print version published in: Cambridge Archaeological Journal 09/2008; 18(03):341 353

Abstract: Franco-Cantabrian cave art continues to be the focus of much speculation but despite the many theories
put forward there has been little progress in explaining the range of perplexing features of this "art." Only by
regarding such wide ranging and anomalous characteristics as central to this debate might some progress as to
derivation be possible. The account presented in this paper will demonstrate how the many "contradictions"
prevailing might provide an important indication as to provenance that can be explained through an understanding
of the shifting nature of visual imagery in the context of the everyday lives of Upper Palaeolithic communities.
This will be based on the notion that the visual world as perceived can be disrupted by certain types of
psychological effects that can be subsequently triggered by particular kinds of stimulus cues and evocative
situations.

Introduction

Explanations as to why Upper Palaeolithic communities came to create cave art have been many and varied with
the latest version emphasising the priority of shamanism in relation to altered states of consciousness (Lewis-
Williams and Dowson 1988; Clottes and Lewis Williams 1996). The latter has, however, been criticised by some
authorities as being either misconceived or unable to account for the many diverse features involved (Hodgson
2006; Helvenston and Bahn 2003; Solomon 1999; Layton 2007). Although Lewis-Williams approach is based on
a universal neurophysiology, it will be demonstrated in this paper that an alternative and more plausible theory can
be constructed that is able to account for many diverse features typical of this art. In this respect, I will
endeavour to demonstrate how a mental phenomenon referred to here as hyperimagery, which is loosely related
but essentially separate from classic hallucinations, can account for the many peculiarities of Palaeoart. The crucial
dynamic is held to be not what the observer perceives as existing beyond or within the cave wall, as Lewis-
Williams suggests, but the interaction obtaining between the perceiver and wall surface itself. In this respect, I
present a theory that radically changes the polemic as to the derivation of cave art. The aim of this approach is to
show how these depictions can be parsimoniously explained by recourse to the everyday experiences of hunter-
gatherer groups in relation to the environment of the late Pleistocene. In this analysis I will present recent research
from neuroscience and related disciplines that will show how these depictions can be understood from the
perspective of how the world is perceived by observers in relation to a range of psychological states. Part 1 will
define hyperimages in the context of neuropsychology whereas Part 2 will show how such images have important
implications for understanding Palaeoart.

Part 1

Visual memory, Perception and Psychological States

The normal visual system, visual memory and imagination

In order to understand how mental images can be relevant to understanding Franco-Cantabrian cave art, it is first
necessary to examine how the visual system processes visual information. In this respect, it has been established
that there is a common neurophysiological mechanism underlying visual representation, whether this involves
perception, imagining, thinking, dreaming or memory (Shepard 1984, Kosslyn 1994, Farah 2000). Crucially, Kirby
and Kosslyn (1992:72) propose that visual memory, as such, evolved out of the perceptual/ recognition system.
What differentiates the various levels of the system depends on whether it is being stimulated by incoming visual
data from the real world, whereupon the accent will be on the early to middle stages, or is being directed from the
need to reflect on events whereby the later stages will come to the fore. This is consistent with the experienced
visual world being a product of interactions between abstract top-down visual memory templates and bottom-up

1
sensory ones, with the former generally more influential in subjective perception. Subjective perception can
therefore be the consequence of dynamic reciprocal interactions among external sensory input, internal object and
scene representations and goal directed attention. This is promoted by top-down biasing of information as a
consequence of familiarity with objects, individual goals, and expectancy that will influence the speed and
accuracy of object awareness (Collerton et al. 2005).
PET scanning studies have demonstrated that some of the same regions of the occipital lobes are indeed
involved in both visual imagery and perception (Kosslyn and Thompson. 2003). Another way of saying this is that
imagined events are produced by some of the same neural architecture dedicated to the perception of actual events.
Visual memory therefore appears to be underwritten by a system that is shared but asymmetrically biased
according to priorities, and one of its properties is to suggest alternatives when confronted with ambiguity, as when
one attempts to disambiguate a camouflaged or indistinct form (Mast 2005). Such an ability would have been
crucial to early humans as an automatic system able to quickly furnish a host of different "suggestions" to the
identity of any ambiguous form that would have led to greater survival rates for those so disposed. The propensity
of the higher end of the visual hierarchy to automatically "suggest" alternatives is a mechanism useful when
danger threatens as well as a device for preparing for actual situations. In this respect, Kosslyn and Shin (1994)
emphasise that imagery selectively facilitates perception, so that if subjects imagine a form they will be more
likely to detect a degraded version of a shape when presented with alternatives. Visual imagination seems to be an
aspect of the visual system closely related to visual memory and one that is most "detachable" from everyday
reality in that it allows possible scenarios to be played out completely disengaged from the real situation. The
obvious benefit of this capacity resides in the opportunity to rehearse situations so that in real circumstances one is
more able to deal with the unexpected. A further useful effect of this would have been the ability to project content
and meaning onto objects that may or may not have any resemblance to things in the real world - an ability with
which humans are particular well equipped even at a young age (Cox 2005). Interestingly, Heider and Simmel
(1944) found that adults, when viewing animated cartoons of a large and small triangle along with a disc portrayed
as moving in and out of a large square, tended to see them as having human characteristics and intentions.

Visual imagery, visual reality and trigger cues

Neisser (1967) proposed that only selected parts of incoming data are represented in memory that provide a starting
point for the reconstruction of a past event. Encoding and remembering are therefore closely related but this can
lead to ambiguities in that present knowledge and needs exert a powerful influence on what is retained (Schacter
1996). These processes are thought to be part of how the brain attempts to impose order on the environment so that
memory does not just depend on the strengths of associations but also on the cues that trigger the initial memory
trace. Furthermore, when a retrieval cue elicits a memory, this is more likely to occur when the circumstances,
especially those concerning emotional arousal, are similar. Interestingly, the trigger cue combines with the original
memory trace to produce a memory experience that is different from that of the original event (Schacter 1996).
The normally functioning visual/memory system may therefore be subject to considerable error depending on
circumstances as is further evidenced by what is termed "The Perky Effect" (Perky 1910). This shows how the
division between imagination and visual reality can be quite fluid. Perky placed observers in a dimly lit room and
asked them to stare at a blank screen some distance away. The observers were then asked to summon up an image
of an object and while preoccupied with this (and still staring at the screen), Perky surreptitiously projected a barely
visible image of an object onto the back of the screen. To simulate the effect of a mental image, the projected
object was agitated slightly to create a shimmering effect. Nearly all the observers remained oblivious to the
attenuated screen projection and thought that what they were seeing was their own internally derived mental image.
Participants not only reported the imagined image being in the same orientation as the projected image but also
unwittingly combined what they saw with what they had imagined e.g., one subject saw a leaf (the projected image)
covered in veins (the mental image) and another a lemon (the projected image) lying on a table (the mental image).
Even after the participants had been informed that what they had witnessed was a real projection, most observers
insisted that they had imagined what was there. This effect has been confirmed by other investigators (Segal and
Fusella 1970, Reeves 1982). These experiments add weight to the notion that our hold on "visual reality" is tenuous
and we are liable to be misled or make errors in certain situations. In other words, imagery can intrude on reality
and reality on imagery. Other studies on the nature of visual imagery have found that there is a loss of high spatial
frequencies thereby reducing the amount of detail thus causing a certain blurring of the remaining outline contours.
(Harvey 1986, Sekuler and Blake 1994 - see especially figures 13.18: 485)

The "normal" visual brain and hyperimagery

2
 Pseudohallucinations are defined as visual images that are not usually mistaken for perception (Davis 1987). The
term pseudohallucination has, however, been criticised as defining too diverse a range of phenomena. I therefore
propose to use the term hyperimagery that better expresses the different causative factors concerned.
Importantly, full blown hallucinations appear to exist in objective space whereas hyperimages seem to reside in
subjective space (Whitlock 1987). Following Sims (2003), Jaspers (1963) and Aggernaes (1972) the main
differences between hallucinations and hyperimagery can be summarised in Table 1.

Hallucinations Hyperimagery

Experience is concrete, tangible, objective, and real Inner eye-pictorial, subjective.

Location in outer objective space. Location in subjective inner space

Good definition Indefinite, incomplete.

Sensory Wealth Neutral or dim elements

Constant Tendency to fade

Independent from volition (experience of passivity) Sometimes experienced as voluntary

Perceptual flavour Ideational flavour

Object exists independent of observer Depends on observer for existence

Experienced in another sensory modality Not experienced in another sensory modality

Table 1. The difference kinds of experiences associated with hallucinations and hyperimagery.

It is obvious from Table 1 that there are important distinctions dividing hallucinations from hyperimagery.
However, there may be a gradation linking the two. In this sense, Krapl Taylor (1979) refers to hyperimagery as a
special type of hallucination. Moreover, hyperimagery may occupy a position along a continuum located between
actual hallucinations, visual imagery and the everyday perception of the real world (Jaspers 1963). It is important
to emphasise, however, that even though hyperimages are not perceived as real they can still have an emotionally
charged dimension (Horowitz 1975). And, as we shall see, although one of the differences between hallucinations
and hyperimagery resides in the fact that, in the latter case, the individual has insight into the fact that the image is
not real, in some circumstances this boundary may be crossed in that there may be a situation where an individual
is uncertain as to whether what is being seen is actually in the minds eye or to do with the object being observed.
In such cases, the image seems to reside in the space between the actual perceived world and the subjectively
induced image. Moreover, as arousal states change this image may also appear to fluctuate between the two
extremes i.e., either appear to be emanating from the real object or surface observed or remain within the limits of
the individuals subjective experience.
Interactions between these various influences are outlined by Horowitz (1975) in stating that internal
representations come to the fore due to several reasons:

1. Relative reduction of external input with no concomitant lowering activity in the representational system.
2. Increase in activity of the representational system without increase in availability of external signals.
3. Augmentation of internal input due to arousal of ideas or feelings secondary to drive states.
4. Reduction of usual levels of inhibition over internal inputs.
5. Alteration of the dynamic between different contingencies permitting internal inputs to gain more

3
 representation over direct perception.

Jaspers (1963) is also careful to make the point that these categories can combine in various ways according to the
mental state of the individual. The relationships between these various categories are summarised in Table 2.

Lack of stimuli from environment Intensification o f external stimuli

leading to greater dependence on leading to greater dependence
subjective imagery on sense perception
More aberrant kinds of subject matter
MENTAL REAL
IMAGERY WORLD
More extreme, abnormal mental states Normal experiences
and anomalous experiences

Increasing ambiguity of external stimuli Decreasing ambiguity of external stimuli

Hallucinations -------- Hyperimagery -------- Visual Imagery -------- Sense Perception
Greater arousal or under arousal Normal arousal levels

Table 2. The various dynamic relationships along a continuum linking sense perception, visual imagery,
hyperimagery and hallucinations with regard to external stimuli and subjective experience. Greater
arousal levels or a more disturbed psychological state would tend to move this continuum further to the
right i.e., in everyday viewing situations an over-aroused person will be more likely to misconstrue
a real object for a subjective image than would a normally functioning individual.

The effects of the realignment of these various factors can be seen in distinct psychological states, examples of
which will now be described. As already indicated, what is remembered from an emotional experience may be
influenced by arousal levels that serve to focus attention on specific aspects of an episode. In such cases, research
has found that people exposed to dangerous events tend to focus on the central themes rather than peripheral
details. This is commonly known as weapon focussing. For example, witnesses to a crime, retain accurate
memories of the particular implement or weapon used. They, nevertheless, have poor memories of the event itself,
including the face of the person holding the weapon that is associated with a narrowing of attention (Schacter
1996). Interestingly, Eagleman (2005) has found that in dangerous situation events appear as if in slow motion
which may explain why particular items are easily transposed to memory.
Focussing-in on an object can similarly lead to hyperimages - especially where the perceptual system has been
subject to increased demands such as where prolonged concentration on particular objects is required when
individuals often report intrusive mental images of the items subsequently appearing. For example, medical
students and laboratory technicians, on learning to count blood cells for long hours through a microscope, later
report seeing intrusive images of cells (Horowitz 1975). Motorists also report the occurrence of visual images of
headlights after long bouts of night-time driving and archaeologists describe recurring images of searched-for
objects or items that have been the focus of intense scrutiny. These experiences, it seems, can be quite startling
because they enter awareness involuntarily and cannot easily be erased. It is as if the image of the object involved
has become so imprinted in neural circuits that it can come to mind in the absence of the real stimulus. This effect
may also be reinforced by the fact that when engrossed in intensely focussed activities attention is distracted from
other concerns. Correspondingly, when an individual endures hunger or thirst, the mental image of the item in
question becomes increasingly intrusive until the need is placated (Gross 1992, Sandford 1936).
Hyperimages can also occur during episodes of fatigue - the opposite of arousal but still a disturbed conscious
state. Those so affected often report seeing objects of various descriptions, invariably animals. For example,
drivers who endure long and monotonous journeys on motorways through fog, snow-storms or at night often report
seeing non-existent animals which they swerve to avoid before the mistake is realised (McFarland and Moore
1957, Whitlock, F. A. 1987). In these cases, the attention to external stimuli is attenuated allowing visual imagery

4
to come to the fore. Lack of sleep, especially over extended periods, can also encourage confusion so that one
object can sometimes be mistaken for another. In extreme cases of sleeplessness, hyperimages can emerge in
response to hints contained in the prevailing stimulus, e.g. spots on a table might be misperceived as insects or a
street sign may be mistaken for a person (Coren 1998).
Although in normal circumstances a memory image fades quickly, images associated with conspicuous events
tend to be retained longer (Vernon 1971) e.g., with the death of a close kin, a complete stranger is often mistaken
for the deceased person (Grimby 1998; Lange et al.1996; Krapl Taylor 1979). As well as in response to
unexpected events, Horowitz (1975) also found intrusive and stimulus repetitive images experienced in over 50%
of cases where individuals had suffered mild to moderate levels of emotional shock, thereby emphasising the
widespread nature of the phenomenon. The attention of such individuals may be easily captured by particular key
stimuli that remind them of a past event in that they tend to be chronically vigilant to the extent that harmless
environmental signals are regarded as a threat.
The factor driving these quasi-hallucinatory episodes is not a psychoactive agent or damage to the brain but
rather the hyper-aroused (or under-aroused) limbic regions that provoke the necessary imagery in higher visual
centres (Joseph 1990). In such cases, the imagery may also be manifest in dreams or nightmares. Anything that
induces over (or under) arousal in these systems may therefore elicit a hyperimage (Manford and Anderman 1998).
The likelihood of a Perky-like effect occurring (mistaking an internally generated image for one based on external
reality) is therefore greatly increased in such cases of distorted arousal.
All of the above examples illustrate how hyperimagery is likely to be experienced by the ordinary population as
part of "normal" experience and need not be confined to exceptional or abnormal groups. Furthermore the division
between what is real and what constitutes a mental image can fluctuate according to the potency of environmental
conditions and the intensity of predisposing psychological factors. The image can thereby appear to float anywhere
in the space between the subjective experience of the observer and the actual world itself. This would, to some
extent, depend on the arousal level of the individual as it is commonly accepted that, when a person is highly
aroused, the perceptual system automatically suggests alternatives. For example, in the case of an alarming event,
the individual will tend to interpret more readily any subsequent stimuli with the slightest resemblance to the
originating cause i.e., detection thresholds will be considerably lowered. In this respect, LeDoux (1994) found that
the subcortical thalamus to amygdala pathway responds rapidly and pr-consciously to potentially threatening
stimuli, especially those involving danger. There are, therefore, two neural pathways for reacting to a dangerous
event, one subcortical (to do with implicit processes) and the other cortical (to do with explicit awareness). The
subcortical route seems to provide only a crude perceptual image of the external world whereas the cortical
pathway embellishes this in the form of a more detailed enhanced representation (Alorda et al 2007). The
subcortical system may trigger an emotional response to a stimulus before conscious recognition ensues, hence
allowing fight or flight mechanisms to be readily tuned this is because not responding to a stimulus involves
more costs than actually responding. The visual system may therefore simply store primitive cues of basic
information that can be verified by a delayed functioning higher recognition system. Thus, the two systems operate
in unison to provide a seamless conscious experience. Interestingly, LeDoux found that these kinds of emotionally
driven memories are difficult, if not impossible, to erase. A consequence of this is that the alerting system
produces a generalised attentional bias to focus on threatening stimuli that tend to persist when activated, as
dangers do not strike in isolation then disappear, but rather tend to linger (hman and Soares 1998). This often
leads to a further entrenchment of arousal levels and an associated increased motivation to seek further threats of
the same order - a mechanism known as adaptive conservatism.
The neuropsychological factors outlined will have been further underwritten by the fact that animals, which
are particularly relevant in the present context, were integral to the evolution of the human brain to the extent that
the encoding of animal forms seems to have become a dedicated domain of the visual cortex (Caramazza and
Mahon, 2006, Farah and Rabinowitz 2003; Capitani et al 2003; Hodgson 2003b, Hodgson 2006) - specifically in
the lateral fusiform gyrus (Caramazza and Mahon 2003; Martin and Chao 2001). A large number of investigations
have confirmed the existence of semantic categories for major items such as animals for the purpose of both
recognition and naming that has become one of the main lines of evidence for constraining conceptual
organization in the human brain (Caramazza and Mahon 2003). Interestingly, the archaeological record (especially
the preoccupation with animals in rock art) seems to confirm what cognitive neuroscience has established.
Moreover, the stimulus properties arising from such a domain will have constituted the basis for more
sophisticated representational abilities (Hodgson and Helvenston 2006). Implicit perceptual processes are also
relevant in this regard, as it has been established that recent and repeated visual exposure to the same object has a
tacit (subconscious) influence on visual memory recall (Hodgson 2003a). This is supported by the fact that studies
of eye movements when viewing complex scenes have found that attention gravitates more to people and animals
than inanimate objects (Yarbus 1967). Such observations are consistent with Gibsons (1979) direct theory of
perception wherein what is perceived is embodied in that the processes of seeing are contingent. In other words,
the categories for the recognition of animals in the cortex are a function of evolutionary imperatives that have an
automatic dimension.

5
 Part 2

Cave Art and Hyperimagery

Hyperimagery and Franco-Cantabrian cave art

How might the present analysis be relevant to understanding the depictions of animals in Franco-Cantabrian
caves? The diverse ways be which hyperimages are induced can be informative as to why there are so many
depictions of animals in this art and the reasons for their representation. Most of the psychological processes and
conditions cited above come together in the caves visited by Upper Palaeolithic people. It has been shown how
hyperimagery of this type can be induced by certain types of experience relating to "normal" interaction with the
world. It is by examining Upper Palaeolithic communities in these terms that we might be in a better position to
assess how hyperimages can be relevant in this context. Fatigue, sleep deprivation, over-concentration on
particular items, anticipation, hunger, and stress related events have been identified as prime causes that can evoke
hyperimagery. Moreover, individuals often find hyperimages so compelling they are occasionally regarded as real
(Sims 2003), which provides an important pointer as to what might have motivated Franco-Cantabrians to treat
these images as special.
It is uncontroversial that such experiences would have been a regular occurrence in the daily lives of Cro-
Magnons. On hunting forays focussed concentration would have been necessary for sustained periods to acquire
the essential protein rich meat and fat products as well as to maintain vigilance in case of marauding predators
(this, however, does not presuppose the much criticised idea of "man the hunter" but merely that all members of
the community would have been involved in hunting activities and a concern for animals to a greater or lesser
extent). As noted, one of the effects of raised and prolonged concentration levels is that the sought-after items
come to be imprinted in neural circuits as a result of increased noradrenergic activation (McGaugh 2006). Mental
fatigue is a hallmark of this level of focussed attention that would have been exacerbated by the extreme physical
exertion involved in tracking game. Hunger, if not starvation (malnutrition was common during the Upper
Palaeolithic [Burenhult 2003]), will have been another important factor (Hodgson 2006) for, as emphasised, lack
of sustenance leads to an image of the food in question spontaneously coming to mind. Furthermore, mental states
play a vital role in how perceptual information is processed to the extent that the hopes, fears and expectations
affect what is perceived (Dror 2005). On entering a cave for what ever reason, the dim conditions meant that the
constraining effect of external stimuli would have been reduced whilst, thanks to recent experiences to do with the
interaction with animals, visual imagery would have come to the fore (bright light improves perception of external
objects and, in this sense, discourages hyperimagery [Collerton et al. 2005]).
Given the daily round of Upper Palaeolithic life, it would have been more than likely that hunter-gatherer
groups did experience such mental aberrations. This would have been reinforced, and has obvious parallels with
the Perky Effect, in that many of the contours of the dimly-lit cave walls simulate animal outlines. That is, there
would have been an ambiguity between the subjectively induced image and the suggestive, undulating contours of
the walls - an ambiguity reinforced by the fluctuating light, and shadows cast, of lamp burners and fires (such light
actually seems to make the walls move thus reinforcing cues for the eliciting of animal forms). This ambiguity will
have been an important motivational factor inducing the visitors to render the mental image permanent.
Horowitz (1975), in relation to hyperimagery, points out that an ambiguous shape (such as a crack or undulating
surface of a cave) can be "illusioned" into either an item of food, if the subject is hungry, or sexual anatomy if in a
state of heightened sexual arousal (which has implications for the vulva iconography in cave art), and so on. It has
been estimated that at least 10-15% of cave art involves the incorporation of natural features that are suggestive of
an animal's anatomy (White 2003). This is probably an underestimation as many of the original rock features on
which the animals were based will have disappeared e.g. natural stains, blotches, protrusions, surface cracks,
undulations, as well as shadows projected on to walls by intervening surfaces, concretions and rock formations. In
fact, Clottes (2007) has recently proposed that the significance of natural features for understanding cave art has
largely been overlooked.
Suggestive features of this order can be regarded as highly potent trigger cues (which parallel and act as a
surrogate and stimulus for similar cues in visual memory) that will have helped encourage and reinforce a visual
system already hyper-sensitised to the detection of animals. The exploitation of natural features that simulate
animals provides a vital clue as to the relevance of hyperimagery and visual memory in cave art. This is reflected
in the fact that on hunting expeditions raised arousal levels served to keep the individual in an alert state so that the
slightest cue suggestive of an animal would have elicited an image of the same or similar. Yet any surface of the
cave, whether flat or undulating, would have provided a suitable "canvas" on which hyperimages may have
potentially been projected so depictions need not be exclusive to areas where natural features of the cave

6
resembled animals. In addition, there are many depictions, although not including suggestive natural features
within the actual contours, are nevertheless often placed in proximity to rock surfaces and features that suggest
animal outlines (see for example Fig. 1 and 2).
The suggestive natural cave features closely resemble abbreviated animal outlines and so replicate incomplete
figures which thereby automatically facilitate and stimulate recognition and visual memory templates. This process
will have been further promoted by an over-stimulated and activated visual system so that natural features that
looked marginally similar to animals would have been enough to incite visual memory or provoke hyperimages. In
other words, Upper Palaeolithic people would have tended to see more animal forms in the natural features of the
cave than a modern individual because Cro-Magnons perceived animals in the evocative cave environment due to
the lower recognition thresholds for triggering such items. Of course, repeated visits to caves over prolonged
periods would have led to a reciprocal process where the original factors that had inspired the depictions of
animals were further compounded by the already existing productions of previous cave artists. In many instances,
the most recent visitors would have remained unaware that the previous depictions had been done by earlier groups
due to loss of contact over time thereby lending a further air of mystery to these sites. Previous depictions
therefore became trigger features in themselves leading to the production of yet further animal outlines a process
that may account for the endurance of these kinds of depictions over such a long period.
It also needs to be remembered that various animals, such as bears, wolves, and cave lions frequented the
caves used by Cro-Magnons, so it would have been necessary to make certain these spaces were uninhabited by
carefully examining the various chambers beforehand. In this case, the visual system would automatically have
been suggesting possible alternatives in the ambiguous and unstable light where there were many false positives
(the suggestive animal-like cave features). Accordingly, on visiting a rock shelter, vigilance levels would have
remained high with the effect that any slight resemblance to an animal outline in the shifting light and shadows
cast onto the undulating walls would have been enough to evoke an image of an animal. And once a hallucinatory-
like image has been experienced in a particular setting it is inclined, in future, to come to mind more easily in that
venue through association (Collerton et al. 2005) which further explains why caves became a favoured location for
the making of Palaeoart. The fact that graphic images are found in the deep more inaccessible parts of caves is
explained by the fact that hyperimagery would have been even more likely to occur in these areas than in the more
accessible chambers. In fact, the strangeness and increased danger associated with these remoter areas would have
further increased arousal and vigilance levels making the appearance of hyperimages more likely
One of the characteristics of hyperimages (and visual imagery) is that they tend to be in the form of a single
representative example of a class or category (Collerton et al. 2005, Vernon, 1971). For animals, this is the typical
viewing profile (see Hodgson 2003b for a discussion), a common trait in Upper Palaeolithic depictions as
Halverson (1991) has demonstrated wherein the contour outline defines the class (see Fig.3). Another
characteristic of hyperimagery is the fact that only the main defining aspects (diagnostic features) of the object are
included in the subjective image in that the features more attended to during interaction with an object tend also to
be those liable to accentuation (Henderson et al. 2001), which, again, matches exactly the characteristics of
Franco-Cantabrian art. For example, the lower extremities of an animal's legs are regularly omitted whereas the
upper diagnostic features are invariably included and drawn in order of priority (Fritz 1999, Hodgson 2003b).
Twisted perspective, where horns, hooves, antlers or tusks might by turned towards the viewer whereas the rest of
the animal remains in profile, is a common aspect of cave art and can be similarly explained by the focussing-in
effect, as in the weapon example mentioned above. The author of the depiction is thereby attending to different
aspects of an animals anatomy and in so doing triggering the typical view for particular anatomical features
antlers, horns and suchlike can potentially cause injury or death (and the hunter could potentially be trampled on
by hooves). This is especially obvious in the twisted horns of the bison in the fallen man scene of Lascaux.
The outline contour of hyperimages has been shown to be incomplete or liable to various levels of fading or
accentuation according to the psychological state and level of awareness of the individual. One of the predominant
aspects of parietal art is not just the concern for the contour but also the abbreviation of this outline in various
ways (the more complete figures of Lascaux are an exception). In fact, the indistinct outline seems to be one of the
commonest traits of parietal art that was intended and not a product of natural deterioration (Ucko and Rosenfeld
1967). Upper Palaeolithic Cave art abounds in such outlines (Ucko and Rosenfeld 1967: 58), the representations of
Chauvet being prime examples with many at Lascaux displaying similar traits as well as the horse panel at Ekain
(Guipzcoa) (Figure 3.) Having said this, there are many depictions that show a concern for the more detailed
aspects of an animal's anatomy and a more defined outline. The more precise outlines can also be explained by
hyperimages in that the latter can often display more definite edges (Jaspers 1963) depending on the intensity of
the experience which is consistent with the view that subjective images can seem quite real if the level of
emotional arousal happens to be high. Otherwise, such portrayals may simply represent those depictions that were
chosen from a previous corpus of outlines for further embellishment for any number of possible reasons (see
below).
Another curious aspect of this art is the disregard for the scale of animals portrayed in the same "scene," (as
Fig. 3 illustrates) which parallels the fact that subjective visual images are subject to size distortions to the extent

7
that an image can range in size from the small to relatively large (Jaspers 1963). A further unusual characteristic
is the juxtaposition of animals not normally seen together, which is all the more surprising given that some
authorities claim that the hunters responsible for cave art were astutely aware of the behaviour of the various fauna
(Clottes 2003). Because hyperimages are unpredictable, intrusive and tend to linger, a species of animal is likely to
have been seen against any number of previously executed portrayals thereby leading to the incongruous
assortment of species.
The many dense superimpositions to be found on cave ceilings seem puzzling to the modern eye but for
Palaeolithic people they were a kind of record of the persistent hyperimages seen against the backdrop of ceilings
and previously painted surfaces (see Fig. 4). The haphazard orientation of these superimposed depictions is
explained by the fact that, for any of the aforementioned reasons, hunters gazing upwards at this ceiling would
have seen hyperimagery from a number of different positions. Another curious aspect of cave art is the rotation of
figures on walls, such as the "vertical rhinoceros" of Chauvet (Clottes 2003). Rotations of this kind are more
frequent on ceilings for reasons already posited and less common on walls due to the fact that the observer's
viewpoint is more likely to be dictated by a common standing posture and therefore a shared line of view.
In sum, the reduced light of the cave with its animal-like undulating contours and the highly charged visual
system of the human visitors will have come together ensuring that hyperimagery of animals would have been
experienced, thus motivating and directing how animals were depicted on cave walls and ceilings. The main
similarities between hyperimagery and Franco-Cantabrian cave art can be summarised as follows:

Incompleteness
Outline contour
Faded quality
Size disparities
Exaggeration and concern for salient features
Single units
Basic category or class

The range of specific psychological factors experienced in the dim and ambiguity-provoking illumination of caves,
in order of precedence, that would have invoked hyperimagery of animals during the Upper Palaeolithic can be
summarised as follows:

1. Prolonged and focussed concentration directed towards various animals

2. Anticipation and expectation
3. Fatigue
4. Sleep deprivation
5. Hunger
6. Fear
7. Stress factors

Hyperimages allow many aspects of Upper Palaeolithic cave art to be understood more clearly that have
previously resisted interpretation. For example, they explain why the depictions are so naturalistic at an early date
in the sense that a hyperimage would have been immediately available for the purpose of portraying fauna and,
when interacting with the suggestive features of the cave, provided an adequate template for producing a
naturalistic depiction. The fact that these representations endured for 20,000 years in virtually the same format is
further corroboration of this explanation in that the same visual image, i.e. the typical viewing profile, will have
remained constant.

Socio-cultural factors and hyperimages.

Upper Palaeolithic peoples must have had a range of experiences in relation to animals including, hunger,
tiredness, focused attention, expectations as well as fear, any one of which would have been enough to elicit a
hyperimage. Also, with fear comes respect to the extent that carnivores can be regarded as positive which is
perfectly consistent with hyperimages. It is precisely because Palaeolithic people would have been wary of and in
competition with large animals that they would also have regarded them with admiration. The carnivores depicted
at Chauvet are thought to portray such a sense of reverence (Robert-Lamblin 2003). To reinforce the point,
threatening animals were only one factor leading to the experience of hyperimages, other factors concerned the
conditions in which Upper Palaeolithic people lived as they interacted with animals on many different levels that
were equally, if not more, important.

8
 The criticism could be raised that humans are, for obvious reasons, more interested in the human figure/face
than animals yet the human form is rare and often poorly drawn in Franco-Cantabrian art. However, as the
population density of humans living in Europe at the time would have been extremely sparse (10,000 in total by
one estimate) the likelihood of coming into contact with other groups of humans would have been extremely low.
Secondly, the suggestive curvilinear contours of the caves are much more similar to animal outlines than the
human figure and, thirdly, as already stated, there would probably have been regular daily contact with animals of
different kinds on a number of levels compared to the more limited interaction with human groups. These points
suggest hyperimages of animals would have been more common than those of humans.
It is highly likely that, because Cro-Magnons will have remained oblivious as to the causes of hyperimagery,
they came to be regarded as special leading to their attempted preservation in a tangible form. Images of this kind
are often referred to as "visions" even in the present day (Reed 1972) and are sometimes associated with haunting
(Lange et al. 1996) which has implications for how Palaeolithic communities might have labelled such
manifestations. The possibility therefore arises that caves may have been concerned with a quest to visit locations
where hyperimages were optimally experienced. And it may also be the case that hyperimages, as realised in
graphic form, were only meant to be seen by certain members of the group because of the special status accorded
to such apparitions.
If hyperimages and their material embodiment did become the focus of socio-cultural factors, the question
remains why was there not much more variation in style and content over such an extended period? One way to
answer this question would be to propose that because the various images existed both in the real world, as
suggestive natural outlines and associated depictions as well as in the minds eye, the interplay between the two
may have had a constraining effect, especially if the images came to be thought of as akin to visions. A powerful
feedback mechanism must have existed between the suggestive wall surfaces of the cave, the daily experiences
with animals, and the visual brain. As visions, the images may also have become proscribed and, in this respect,
integrated into a broader mythic narrative that remains archaeologically invisible. The fallen man scene from
Lascaux, as one of the few examples of a pictorial narrative, tentatively suggests this to be the case. Theranthropes
may similarly be cited as evidence of mythic or socio-cultural leanings. Modern hunter-gatherer groups have a
complex relationships with animals that is usually associated with some all embracing narrative whereby the
boundaries between humans and animals are held to be fluid (Fausto 2007). The animals portrayed in Franco-
Cantabrian cave art may, in a way that still remains obscure, be part of such a complex narrative unrelated to
shamanism.

The explanatory scope of hyperimages.

Taken together with the fact that animals continued to be a crucial aspect of the everyday lives of Cro-Magnons, it
is incontrovertible that the profiles of fauna often came to the mind's eye of such individuals thus explaining why
animals were so pervasive in this "art." To summarise, the range of phenomena of cave art that can be explained
by hyperimages are as follows:

The preoccupation with animal outlines for 20,000 years

Apparent naturalism at an early date
The inclusion of undulating wall surfaces and fissures and other features as part of animal form
Superimpositions
Haphazard placement of images
Location of depictions on ceilings and in deep recesses
Caves as favoured location
The universality of the typical viewing profile
Size differentials of animals
Various orientations of animals on walls and ceilings
Indistinct, incomplete outlines
Animals portrayed together that do not co-exist in reality
Twisted perspective

Conclusion

Franco-Cantabrian caves served as a particularly favourable location that promoted and encouraged the experience
of hyperimagery with many natural features serving as trigger cues for visual memory. Upper Palaeolithic groups
visiting these caves will have engaged in demanding situations on expeditions involving highly focussed and

9
 prolonged concentration that led to an accentuation of visual imagery at the expense of externally perceived
stimuli. This may have been further reinforced by hunger or experiences related to the practicalities of tracking
game. The classic range of experiences that provoke hyperimagery appear to have been common to those
inhabiting the Upper Palaeolithic refugium, namely prolonged focussed attention, anticipation, starvation, fatigue,
sleep deprivation as well as various kinds of psychological stress. Because Upper Palaeolithic people would have
been unaware of the reasons for the existence hyperimagery, it is probable that this imagery came to be regarded as
special. Indeed, the dimly lit cave environment may have been particularly associated with "visions" of animals.
The fact that the depictions of fauna show many similarities with hyperimagery strongly suggests that this type of
imagery is responsible for many aspects of Franco-Cantabrian cave art. Unlike many previous explanations, this
account is firmly based on the regular activities of Upper Palaeolithic people as they attempted to survive during
the fluctuating climate of the Ice Age - conditions that directly influenced their psychological state that directly
influenced how parietal art was produced.1

1
One of the advantages of this theory is that it can be potentially corroborated by evidence from modern-day populations . For example, an
experiment could be devised whereby subjects might be given an identification task involving prolonged concentration of animal pictures
thereby simulating the perceptual activity with which Upper Palaeolithic hunters would have been concerned in locating and searching for
animals. The targeted group could then be asked to enter a cave (with suggestive contours) and asked to record what they saw. A control group
would simply follow their normal routine and their comments on what they witnessed in the cave also registered. The prediction is that the
target group would tend to see more animal forms than the control group. This procedure could be made more authentic by giving the focal
group adrenaline to induce arousal that would predict the seeing of an even greater number of animals as a consequence. Alternatively, the
target group could be deprived of sleep and similarly admitted to a cave and their responses recorded which could then be compared to the
reports of a normal-sleep control group. Although these experiments could not prove the authenticity of hyperimages as applied to Palaeoart, it
would at least be informative.

Glossary

Amygdala
One of two small, almond-shaped areas of grey matter that are part of the limbic located in the temporal lobes of the
cerebral hemispheres concerned with the memory of emotional responses.

Arousal
Is a physiological and psychological raised state of wakefulness. It involves the activation of the reticular activating
system in the brain stem, autonomic nervous system and endocrine system, leading to increased heart rate and blood
pressure and a condition of sensory alertness, mobility and readiness to respond. Four major systems originating in
the brainstem, with connections extending throughout the cortex, are based on the brain's neurotransmitters,
acetylcholine, norepinephrine, dopamine, and serotonin. When these systems are in action, the receiving neural
areas become sensitive and responsive to incoming signals.

Emotion
A complex reaction concerning experiential, behavioural, and physiological factors whereby an individual strives to
deal with personally significant matters and events that arises without conscious effort and can be positive or
negative depending on circumstances.

Fusiform Gyrus
The fusiform gyrus is situated in the lower area of the temporal lobe (on each side of the inferior cerebral cortex)
and also known as the occipitotemporal cortex and is implicated in visual memory.

Limbic
The limbic system is a part of the human brain concerned with emotion, motivation, and emotional memory. It
influences the formation of memory by integrating emotional states with stored memories of physical sensations.

Noradrenergic (noradrenaline) - Also known as norepinephrine.

A stress hormone that affects parts of the human brain where attention and responses to actions are controlled.
Along with epinephrine, noadrenaline underlies the fight-or-flight response, directly increasing heart rate, triggering
the release of glucose from energy stores, and increasing skeletal muscle readiness

Subcortical - beneath the cerebral cortex.

Thalamaus - relays sensory impulses to the cerebral cortex.

10
References

Aggernaes, A. 1972. The experienced reality of hallucinations and other psychological phenomena. Acta
Psychiatrica Scandinavica. 48, 220-38.

Alorda, C., Serrano-Pedraza, I., Campos-Bueno, J, J., Sierra-Vazquez, V. and Montoya, P. 2007. Low spatial
frequency filtering modulates early brain processing of affective complex pictures. Neuropsychologia. 45, 3223-
3233.

Burenhult, G.. 2003. Modern People in Africa and Europe, in (ed.) Burenhult, G. People of the Past, San Francisco
(CA): Fog City Press. 77-95.

Capitani, E., Laiacona, M., Mahon, B. & Caramazza, A. 2003. What are the facts of semantic category specific
deficits? A critical review of the clinical evidence. Cognitive Neuropsychology. 20 (3), 213-261.

Caramazza, A. & Mahon, B. Z. 2006. The organisation of conceptual knowledge in the brain: The future's past and
some future directions. Cognitive Neuropsychology. 23 (1), 13-38.

Caramazza, A. & Mahon, B. Z. 2006. The Organization of Conceptual Knowledge: The evidence from category
specific semantic deficits. Trends in Cognitive Sciences. 7 (8), 354-361.

Clottes, J. 2003. Return to Chauvet Cave. Excavating the Birthplace of Art: The First Full Report. London:
Thames and Hudson.

Clottes J. & Lewis-Williams J. D. 1996, Les chamanes de la prhistoire: transe et magie dans les grottes ornes,
Paris, Le Seuil.

Clottes, J. 2007. New Discoveries at Niaux Cave in the French Pyrenees. in, Exploring the Mind of Ancient Man :
Festschrift to Robert G. Bednarik, editor P. C. Reddy. New Delhi: Research India Press, 281-291

Collerton, D., Perry, E. & Mckeith, I. 2005. Why people see things that are not there: A novel Perception and
Attention Deficit model for recurrent complex visual hallucinations. Behaviorial and Brian Sciences. 28, 737-94

Coren, S. 1998. Sleep Deprivation, Psychosis and Mental Efficiency. Vol. XV. (3), Psychiatric Times. Available
at: http://www.psychiatrictimes.com/p980301b.html

Cox, M. V. 2005. The Pictorial World of the Child. Cambridge University Press.

Davis. D. R. 1987. Hallucination, in, The Oxford Companion to The Mind, Editor R. L. Gregory. Oxford: Oxford
University Press, 299-300.

Dror, H. E. 2005. Perception is far from perfection: The role of the brain and mind in constructing realities.
Behaviorial and Brian Sciences. 28, 763.

Eagleman, D. M, Tse, P. U, Buonomano, D, Janssen, P, Nobre, A. C. & Holcombe, A. O. 2005. Time and the
Brain : How Subjective Time Relates to Neural Time. The Journal of Neuroscience, 25 (45), 10369-71.

Farah, M. J. 2000. The Cognitive Neuroscience of Vision. Malden, Mass.: Blackwell.

Fausto, C. Feasting on People: Eating Animals and Humans in Amazonia. Current Anthropology. 48 (4), 497-530.

Fritz, C. 1999. Towards the reconstruction of Magdalenian artistic techniques: the contribution of microscopic
analysis of mobiliary art. Cambridge Archaeological Journal 9, 189-208.

Gibson, J. J. 1979. The Ecological Approach to Visual Perception. Boston: Houghton Mifflin.

Grimby, A. 1998. Hallucinations following the loss of a spouse: Common and normal events among the elderly.
Journal of Clinical Geropsychiatry. 4, 65-74.

11
Gross, R. D. 1992. Psychology - The Science of Mind and Behaviour. London: Hodder & Stoughton.

Halverson, J. 1991. Paleolithic Art and Cognition. The Journal of Psychology. 126 (3), 221-36.

Harvey, L. O. Jr. 1986. Visual memory: what is remembered? in, Human memory and cognitive capabilities, eds.
F. Klix & H. Hagendorf. The Hague: Elsevier, 173-87.

Heider, F. & Simmel, M. 1944. An experimental study of apparent behaviour. American Journal of Psychology.
57, 243-249.

Helvenston, P. A., and Bahn, P.G. 2003. Testing the "Three Stages of Trance" Model. Cambridge Archaeological
Journal 13 (2), 213-224.

Henderson, J. M, Falk, R, Minur, S, Dyer, F. C. & Mahedevan, S. 2001. Gaze control for face learning and
recognition by humans and machines, in From fragments to objects. Segmentation and grouping in vision, eds. T.
F. Shipley and P. J. Kellman. London: Elsevier, 463-81

Hodgson, D. 2000. Shamanism, Phosphenes, and Early Art: An Alternative Synthesis. Current Anthropology.
41 (5):866-73.

Hodgson, D. 2003a. Seeing the "Unseen": Fragmented Cues and the Implicit in Palaeolithic Art. Cambridge
Archaeological Journal. 13 (1), 97-106.

Hodgson, D. 2003b. The biological foundations of Upper Palaeolithic art: stimulus, percept and representational
imperatives. Rock Art Research. 20 (1), 322.

Hodgson, D. 2006. Altered States of Consciousness and Palaeoart: an Alternative Neurovisual Explanation.
Cambridge Archaeological Journal. 16 (1), 27-37.

Hodgson, D. & Helvenston, P. A. 2006. The Emergence of the Representation of Animals in Palaeoart: Insights
from evolution and the cognitive, limbic and visual systems of the human brain. Rock Art Research. 23 (1), 3-40.

Horowitz, M. J. 1975. Hallucinations: An information-processing approach, in Hallucinations - Behaviour,

Experience and Theory, in editors R. K. Siegel L. & J. West. New York: John Wiley and Sons, 163-95.

Jaspers, K. 1963. General Psychopathology. (Trans. by J. Hoenig and M. W. Hamilton) Manchester, (U.K.):
Manchester University Press.

Joseph, R. 1990. Neuropsychology, Neuropsychiatry, and Behavioral Neurology. New York: Plenum.

Kirby, K. N. & Kosslyn. S, M. 1992. Thinking Visually, in Understanding Vision, editor G. W. Humphreys.
Cambridge, Mass: Blackwell, 71-85.

Kosslyn, S. M. 1994 Image and Brain. Cambridge, Mass: Bradford (MIT press).

Kosslyn S. M. & Shin L. M. 1994. Visual Mental Images in the Brain: Current Issues," in The Neuropsychology of
High Level Vision. Collected Tutorial Essays, eds. M. J. Farah. and G. Ratcliff. Hillsdale, (NJ): Lawrence Erlbaum
Associates. 269-96.

Kosslyn, S. M. & Thompson, W. L. 2003. When is early visual cortex activated during visual mental imagery?
Psychological Bulletin. 129, 723-46.

Krapl Taylor, F. 1979. Psychopathology - Its Causes and Symptoms. Sunbury-on-Thames: Quartermaine House.

Lange, R, Houran, J, Harte, T. M. & Havens, R. A. 1996. Contextual mediation of perception in hauntings and
poltergeist-like experiences. Perceptual and Motor Skills. 82, 755-62.

Layton. R. 2007. Is shamanism a unitary phenomenon (and can it help elucidate rock art)? Proceedings of
Archaeological Invisibility and Forgotten Knowledge. World Archaeological Congress. (In press).

12
LeDoux, J. E. 1994. Emotion, memory and the brain. Scientific American (June): 32-9.

Lewis-Williams, J. D. & Dowson, T. A. 1988. The signs of all times: entoptic phenomena in Upper Palaeolithic
art. Current Anthropology 24, 201-245.

Manford, M. & Anderman, F. 1998. Complex visual hallucinations: Clinical and neurobiological insights. Brain
121, 1819-40.

Martin , A. and Chao, L.L. 2001. Semantic memory and the brain: structure and processes. Current Opinion in
Neurobiology. 11, 194-201.

Mast, F. W. 2005. Mental Images: Always present, never there. Behavioral and Brain Sciences. 28, 737-94.

Mcgaugh, J. L. 2006. Make mild moments memorable: add a little arousal. Trends in Cognitive Sciences. 10, 345-
47.

McFarland, R. A. & Moore, R. D. 1957. Human factors in highway safety: a review and evaluation. New England
Journal of Medicine. 256, 792-99.

Neisser, U. 1967. Cognitive psychology. New York: Appleton-Century-Crofts.

hman, A. & Soares. J. J. F. 1998. Emotional conditioning to masked stimuli: expectancies for aversive outcomes
following nonrecognized fear-relevant stimuli. Journal of Experimental Psychology: General. 127 (1): 69-82.

Perky, C. W. 1910. An experimental study of imagination. American Journal of Psychology, 21, 422-52.

Farah, M. J. & Rabinowitz, C. 2003. Genetic and Environmental Influences on the Organisation of Semantic
Memory in the Brain: Is Living Things an Innate Category? Cognitive Neuropscychology. 20 (3), 401-408.

Reed, G. 1972. The Psychology of Anomalous Experience - A Cognitive Approach. London: Hutchinson University
Press.

Reeves, A. 1982. Letter to the editors. Vision Research. 22, 711.

Robert-Lamblin, J. 2003. An Anthropological View, in Return to Chauvet Cave, J. Clottes (ed.) London: Thames
and Hudson, 200-207.

Schacter, D. L. 1996. Searching for Memory: the Brain, the Mind, and the Past. New York (NY): Basic Books.

Sandford. R. N. 1936. The effects of abstinence from food on imaginal processes. Journal of Psychology. 2, 129-
36.

Segal, S. J. & Fusella, V. 1970. Influence of imagined pictures and sounds on detection of visual and auditory
signals. Journal of Experimental Psychology. 83, 458-64.

Sekuler, R. & Blake, R. 1994 (3rd edition) Perception. New York: McGraw Hill.

Sharpe, K. and Van Gelder, L. 2007. Symbols of a young girl from the Upper Palaeolithic: A tectiform in
Rouffignac Cave, France. (In process).

Shepard, R. N. 1984. Ecological Constraints on Internal Representation: Resonant Kinematics of Perceiving,

Imagining, Thinking and Dreaming. Psychological Review. 91 (4), 417-47.

Sims, A. C. P. 2003. Symptoms in the Mind. Saunder: Elsevier Science.

Solomon, A. 1999. Meanings, models and minds: a reply to Lewis-Williams. South African Archaeological
Bulletin. 54, 51-60.

Ucko, P. J. & Rosenfeld, A. 1967. Palaeolithic Cave Art. London: Weidenfeld and Nicolson.

13
 Vernon. M. D. 1971. The Psychology of Perception. Harmondsworth: Penguin.

White, R. 2003. Prehistoric Art - the symbolic journey of humankind. New York: Harry N. Abrams.

Whitlock, F. A. 1987. Pseudohallucination, in The Oxford Companion to the Mind, editor R. L. Gregory. Oxford:
Oxford University Press. 648.

Yarbus, A. L. 1967. Eye movements and vision. New York: Plenum Press.

14

View publication stats