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

net/publication/313730067

Neuropsychology

Chapter · December 2017

DOI: 10.1016/B978-0-12-809324-5.02465-2

CITATIONS READS

0 722

1 author:

Giovanni Berlucchi
University of Verona
13 PUBLICATIONS   60 CITATIONS   

SEE PROFILE

All content following this page was uploaded by Giovanni Berlucchi on 28 August 2018.

The user has requested enhancement of the downloaded file.

 Journal of Applied Research in Memory and Cognition 6 (2017) 22–26

Contents lists available at ScienceDirect

Journal of Applied Research in Memory and Cognition

journal homepage: www.elsevier.com/locate/jarmac

Radical Cognitivism? Distinguishing Behavior from Thought夽

Colin M. MacLeod ∗ and Evan F. Risko

University of Waterloo, Canada

Keywords: Functional explanation, Cognitive explanation, Applied psychology

De Houwer, Hughes, and Barnes-Holmes offer up what functionalism. As Watson (p. 25) puts it “behaviorism is the
appears to be a simple recommendation for the field: that only consistent and logical functionalism.” In contrast, classical
we should distinguish—and even separate—our descriptions functionalism was a reaction to structuralism, which defined
of the phenomena that we investigate from our theoretical the proper domain of psychology as the study of consciousness
accounts of those phenomena. They characterize this as a distinc- and the mental structures that composed conscious experience,
tion between functional explanation and cognitive explanation, and which rested heavily on the method of introspection
respectively. And they maintain that more assiduous applica- (Titchener, 1929, p. 43). Functionalism sought to discard the
tion of this distinction would be of value not only for basic method—introspection—and instead emphasized dynamic (and
research but also for applied research, the focus of this arti- physiological) adaptation to the environment for both behavior
cle. In this commentary, we consider several aspects of this and “mental life,” but still held the study of consciousness to
proposal, including its emphasis on the stimulus-environment- be central to the mission. It thus appears that the functionalism
response level of description, the implications for the labeling of of De Houwer et al. is more akin to the behaviorism of Watson,
phenomena, the rationale for separating description from expla- keeping in mind, of course, that the cognitive level of explana-
nation, and the potential consequences for the application to tion is superimposed on the functional explanation. De Houwer
basic research. On some of these, we find ourselves in agree- et al., of course, are clear on this commitment noting that the
ment; on others, we respectfully disagree. “functional approach. . .can be linked to behaviorism, at least
certain forms of behaviorism such as radical behaviorism” (p. 1).
The Functional Level of Explanation This commitment is important as it suggests a deeper incompat-
ibility between the functional and cognitive perspectives than is
De Houwer et al. (p. 2) hold that “Within the functional level maybe reflected in De Houwer et al.’s characterization of it as an
of explanation, behavior is explained in terms of the (current or “illusory barrier” (p. 9). This particular issue with the functional-
past) environment and the way organisms interact with the envi- cognitive approach has been cogently argued in a separate
ronment.” This echoes an earlier characterization by De Houwer comment on the approach (see Proctor & Urcuioli, 2016).
(2011, p. 204), that “A cornerstone of the functional approach We are entirely in agreement with De Houwer et al. that our
in psychology is the practice of defining behavioral effects first job as psychologists should be to carefully and thoroughly
exclusively in terms of elements in the environment.” Interest- analyze and describe the phenomena that we study. Indeed,
ingly, there is also a more distant echo: “The psychology which without doing so, we would argue that we would have nothing
I should attempt to build up would take as a starting point, first, to explain. And there is no question that it is misguided to
the observable fact that organisms, man and animal alike, do equate behavioral effects with mental processes in a kind of
adjust themselves to their environment” (Watson, 1913/1994, p. one-to-one mapping, a point that has been made in the past
250). In this sense, the functional approach that De Houwer et al. (e.g., Jacoby, 1991). It is a virtual certainty that phenomena, or
champion is more aligned with behaviorism than with classical tasks, can never be process-pure: “problems interpreting task

Author Note
Preparation of this commentary was supported by discovery grants from the ∗ Correspondence concerning this article should be addressed to Colin M.

Natural Sciences and Engineering Research Council of Canada to each of the MacLeod, Department of Psychology, University of Waterloo, Waterloo, Ontario
authors. N2L 3G1, Canada. Contact: [email protected]
夽 Please note that this paper was handled by the current editorial team of

JARMAC.
 RADICAL COGNITIVISM 23

dissociations have arisen from equating particular processes by its (more functional) description, that is, a detection cost at
with particular tasks and then treating those tasks as if they the location of the previous cue. Such a characterization leaves
provide pure measures of those processes” (Jacoby, 1991, p. potential deeper cognitive accounts more open; of course, it is
513). It is not surprising, therefore, that a great deal of the also more a description than a name. Yet, as Berlucchi laments,
creativity and effort in research on cognitive phenomena is the theory-driven name of the phenomenon lives on.
leveled at trying to “clean up” tasks to eliminate as many Carefully considering the organism–environment interaction
processes as possible other than the one that is the target of in the De Houwer et al. version of a functional approach would,
primary interest. Still, we know that we never reach the “holy we believe, help to preclude naming a phenomenon using the
grail” of isolating one and only one process. Thus, De Houwer initial cognitive theoretical explanation for that phenomenon.
et al.’s reminder that we need to remain vigilant against equating Others have made this same argument which led MacLeod et al.
behavioral effects with mental processes is certainly welcome. (2003) to suggest that we adopt the language of cost and benefit
One of the two illustrations that De Houwer (2011) uses to to avoid slipping from the outset into more cognitively loaded
show why functional and cognitive levels should be separated is terms such as inhibition and facilitation. Clearly, separate termi-
the case of negative priming. Negative priming is the situation nology for phenomena and their cognitive explanations would,
where a target item on a current trial was an ignored item on as De Houwer et al. maintain, help to avoid this trap. We cer-
a prior trial (e.g., Fox, 1995; Tipper, 2001) and the result is tainly agree that “different researchers use the same concepts at
slowed processing on the current trial, relative to an unrepeated multiple levels of explanation” (p. 9), and that this can be very
control. The cost evident in the phenomenon has predominantly confusing and can lead subsequent researchers astray.
been explained by inhibition—that to be ignored, the prior trial
item must have been suppressed, with that suppression carrying
over to slow the processing on the current trial. This was an
appealing and intuitive account, such that negative priming Separating Description from Explanation
came to be treated as a measure of inhibition: The theory “Defining behavioral effects in purely functional terms not
became grafted to the phenomenon. But over time, another only maximizes theoretical freedom but also promotes cumu-
theory—the episodic retrieval account—gained in currency, lative science” (De Houwer et al., p. 4). This sounds like an
explaining the same cost as due to the response appropriate for unassailable truth . . . but is it? Most scientists would quickly note
the current trial conflicting with that retrieved from the prior that research should test theories, not just explore phenomena,
trial. It was not easy for the retrieval account to gain traction so if phenomena must take precedence, that represents a prob-
given the prevalence of the inhibition account, although the lem. We are of the view that the stringent testing of theory
inhibition account is probably the less-supported account now should most often directly motivate empirical research. But there
(see MacLeod, Dodd, Sheard, Wilson, & Bibi, 2003; Mayr & must also be room for discovery, in the sense of finding a new
Buchner, 2015). Not keeping the phenomenon and the theory phenomenon and pursuing it empirically for a while until its
distinct from each other may well have hampered progress, boundaries provide a basis for theorizing. Relatedly, there should
consistent with the argument of De Houwer et al. be room in psychological science for rich descriptions of human
behavior, particularly as it occurs in relatively natural condi-
tions (Kingstone, Smilek, & Eastwood, 2008; Neisser, 1978;
The Nominal Fallacy
Risko, Richardson, & Kingstone, 2016; Tunnell, 1977). This is
Philosophers refer to the nominal fallacy as the belief that not, however, coincident with simply exploring a phenomenon
something has been explained when it has been given a name. for its own sake and never stopping to link its explanation to
This, certainly, is a seductive error, one that likely is invited broader theory. So yes, theoretical freedom is greater if not
by integrating the description (functional) with the explanation pinned down from the outset, but science ultimately hinges on
(cognitive). Negative priming borders on this, with “negative” the development of good theories.
suggesting “below-zero suppression,” but there are better There is another side to this claim, that of promoting
examples. A perfect illustration is the phenomenon know as cumulative science, and it is here that a problem can arise. A
“inhibition of return,” in which, after a short delay, people are scientist can become too enmeshed in a phenomenon, leading
slower to return to a recently cued item or location even though to the “what hasn’t been done yet?” mentality. Dustbowl
the cue has no validity (Posner & Cohen, 1984; Posner, Rafal, empiricism—collecting data for their own sake without regard
Choate, & Vaughan, 1985; see Lupiáñez, Klein, & Bartolomeo, to theory—will not in the end advance science to the same
2006). Why? The longstanding answer is because when nothing extent as theory-driven research will (although mining the same
was initially found at the cued location, attention prefers not to strip repeatedly may advance a prolific scientist’s career). Too
go back there, inhibiting the location, much as a well-adapted much focus on the functional side would seem to enhance the
bird would not return to a cache site that has already been emp- likelihood of this problem, as has certainly happened in the
tied. This inhibition account may eventually be proven correct, history of psychology. An example might be the verbal learning
but that is not a given and there are certainly alternative accounts tradition (see Hall, 1971) that predated the modern study
(see, e.g., Berlucchi, 2006; Martín-Arévalo, Kingstone, & of memory, wherein simply relating stimulus and response
Lupiáñez, 2013). Thus, Martín-Arévalo et al. argue that instead conditions seemed like the goal in itself, but there was a serious
of “inhibition of return” the phenomenon should be referred to shortcoming in the development of understanding. We would
 RADICAL COGNITIVISM 24

not want to see this problem exacerbated by separation of the This argument reminded us of the Alan Newell’s famous admon-
functional and cognitive approaches. itory paper “You Can’t Play Twenty Questions with Nature and
Win. . .” and his encouragement to build more integrative, uni-
On Analytic Abstractive General Functional Principles fied theories of cognition (Newell, 1973, 1994). There is wisdom
in that sentiment (see Meiser, 2011, for recent discussion). This
An important contribution of the paper was the description was not, we think, the kind of wisdom that De Houwer et al.
of and argument for the utility of what De Houwer et al. refer had in mind, but we do believe that applied psychology benefits
to as analytic abstractive functional research. This approach from the search for general principles of human cognition (and
consists of attempts to discover general principles that cut across also that the latter search can benefit from applied research).
particular environment–behavior relations (e.g., reinforcement). While allowing that cognitive research can be too “effect-
In cognitive psychology, these general principles would be tied centric,” it is also important to point out that there exist numerous
to specific mediating mental mechanisms. But this apparently examples to the contrary (e.g., research on working memory
clean distinction seemed to us to break down at times. Perhaps and attentional control; Engle, 2002; conflict monitoring and
the signal violation of the separation occurred for us on p. 10, cognitive control; Botvinick, Cohen, & Carter, 2004; executive
where we are told that “from a functional perspective, implicit functions; Miyake et al., 2000; dual processing theory; Evans &
memory effects are behavioral effects of past events that occur Stanovich, 2013). In addition, whereas De Houwer et al. suggest
under conditions of automaticity.” Surely automaticity is not a that “there are few if any examples of [Cognitive Inspired Ana-
functional idea! In addition, we are presented with an example lytic Abstractive] research in the applied literature” (p. 6), there
of a novel contribution of such an analytic abstractive approach are numerous examples wherein researchers have taken general
on p. 5 where the Stroop effect and Simon effect are claimed cognitive principles (rather than general functional principles)
to be instances of the general functional principle of stimulus and used them successfully in applied settings (e.g., cogni-
control. The notion of “stimulus driven” (versus goal-driven), tive load theories in multimedia learning; Mayer & Moreno,
“bottom-up” (versus top-down) control has a long history in 2003; multiple resource theory in multi-tasking environments;
cognitive psychology: How are these constructs meaningfully Wickens, 2008). Of course, this is not to imply that such trans-
different? Furthermore, whereas Stroop and Simon effects lation is always smooth and that there is no room for change in
certainly share surface features, there is disagreement over our standard operating procedure with the goal of increasing the
whether the cognitive mechanisms underlying them are the likelihood that this translation is successful (e.g., taking more
same (e.g., Egner, Delano, & Hirsch, 2007; Kornblum, 1994; seriously concerns about ecological validity; Kingstone et al.,
Simon & Berbaum, 1990). 2008; Neisser, 1978; Risko et al., 2016).
At times, then, the specifics of the current argument for adopt- The translation from basic research in cognition (e.g., prin-
ing an analytic abstractive functional approach escaped us. This ciples of memory function) to specific applied contexts (e.g.,
may be our own failing as De Houwer et al. point out after provid- eyewitness identification) seemingly requires that research be
ing a description of dot-probe effects using the general functional conducted in De Houwer et al.’s Cognitively Inspired Effect
principle of stimulus control: “We realize that for a cognitive Centric research cell. This is arguably where the authors would
researcher, this analysis is likely to be perceived as an awk- locate much of the research conducted under the banner of
ward redescription. . .It is awkward because it is coined in terms applied cognitive psychology (and most of the work published
unfamiliar to many cognitively-inspired researchers” (p. 8). in this journal). But De Houwer et al.’s take on this cell seems
Although we are happy to take some of the responsibility, in primarily critical. Again they note that such research is at risk
the spirit of genuinely trying to understand the proposal in front of using behavioral effects as proxies for mental processes but
of us, we hope that the future holds less awkward explanations. add that such research also risks failing because the cognitive
model on which it might be based might not be “up for the job.”
Implications for Applied Research This latter concern, however, seems as much a problem for De
Houwer et al.’s preferred Cognitively Inspired Analytic Abstrac-
The primary impetus for the De Houwer et al. article would tive research as it does for the Cognitively Inspired Effect Centric
seem to be with respect to applied psychology, given that the research. Furthermore, while adopting a functional-cognitive
basic functional–cognitive separation argument has previously framework might help us avoid using behavioral effects as
been made elsewhere (De Houwer, 2011; De Houwer, Barnes- proxies for mental processes, it is unclear how it would help
Holmes, & Moor, 2013; De Houwer, Gawronski, & Barnes- us get the cognitive model right, arguably the deeper prob-
Holmes, 2013). The argument here is that applied researchers lem to solve. Maybe we need a more radical cognitivism rather
will benefit from a shared language that is functionally based. than radical behaviorism? All that said, it is early days for the
Despite the extended discussion, we did not see how the 2 × 2 functional-cognitive framework and its impact remains to be
taxonomy dividing applied researchers into four types would seen. We certainly welcome the effort to increment the likeli-
yield such a benefit—beyond serving as a ready reminder to be hood that basic research in cognition can have a positive impact
cautious in using phenomena as mental proxies—nor did we see on our day-to-day lives.
how this taxonomy was beneficial in itself. One feature of this With respect to the functional–cognitive separation and its
taxonomy that did strike a chord with us was the concerns raised relation to applied psychology, we are reminded of one of the
about a psychological science that becomes too “effect-centric.” most famous statements in all of psychology—Lewin’s (1951,
 RADICAL COGNITIVISM 25

p. 169) “There is nothing so practical as a good theory.” As References

Greenwald (2012) notes, a “good” theory in the Lewinian sense
Berlucchi, G. (2006). Inhibition of return: A phenomenon in search of
will go beyond capturing a laboratory situation, extending to the
a mechanism and a better name. Cognitive Neuropsychology, 23,
world beyond the lab where it then can lead to worthwhile appli-
1065–1074.
cations. Cognitive theory certainly has done that, as applications Botvinick, M. M., Cohen, J. D., & Carter, C. S. (2004). Conflict moni-
to education illustrate (see, e.g., Dunlosky, Rawson, Marsh, toring and anterior cingulate cortex: An update. Trends in Cognitive
Nathan, & Willingham, 2013). Moreover, as Greenwald also Sciences, 8, 539–546.
points out, titling his article “There is Nothing so Theoretical as De Houwer, J. (2011). Why the cognitive approach in psychology would
a Good Method” in tribute to Lewin, cognitive theory has often profit from a functional approach and vice versa. Perspectives on
brought about the development of new methods. Greenwald pro- Psychological Science, 6, 202–209.
vides the examples of the additive factors method of Sternberg De Houwer, J., Barnes-Holmes, D., & Moor, A. (2013). What is learn-
(1969) and the process dissociation method of Jacoby (1991). ing? On the nature and merits of a functional definition of learning.
Methods are presumably at the functional level, so clearly the Psychonomic Bulletin & Review, 20, 631–642.
De Houwer, J., Gawronski, B., & Barnes-Holmes, D. (2013). A
field has benefitted from “cross-pollination” between the two
functional-cognitive framework for attitude research. European
levels.
Review of Social Psychology, 24, 252–287.
Finally, we see the introduction of the term “psychological Dunlosky, J., Rawson, K. A., Marsh, E. J., Nathan, M. J., & Will-
engineering” as more likely to be confusing than helpful, given ingham, D. T. (2013). Improving students’ learning with effective
the quite longstanding existence of a field called engineering learning techniques: Promising directions from cognitive and educa-
psychology, which aims to use what we know about human tional psychology. Psychological Science in the Public Interest, 14,
behavior to inform how technology and systems are designed 4–58.
and controlled. Indeed, the authors note this potential confu- Egner, T., Delano, M., & Hirsch, J. (2007). Separate conflict-specific
sion in a footnote themselves, foreseeing the confusion the term cognitive control mechanisms in the human brain. Neuroimage, 35,
would likely create. Put simply, we see nothing wrong with the 940–948.
existing term, applied psychology. Furthermore the 2 × 2 taxon- Engle, R. W. (2002). Working memory capacity as executive attention.
Current Directions in Psychological Science, 11, 19–23.
omy offered for applied psychology by De Houwer et al. seems
Evans, J. S. B., & Stanovich, K. E. (2013). Dual-process theories of
unabashedly cognitive and, while this resonates with our own
higher cognition: Advancing the debate. Perspectives on Psycho-
biases, we wonder whether applied psychologists of different logical Science, 8, 223–241.
stripes might be as enthusiastic. Fox, E. (1995). Negative priming from ignored distractors in visual
selection: A review. Psychonomic Bulletin & Review, 2, 145–173.
Greenwald, A. G. (2012). There is nothing so theoretical as a good
In Conclusion method. Perspectives on Psychological Science, 7, 99–108.
Hall, J. F. (1971). Verbal learning and retention. Philadelphia: Lippin-
To summarize, we agree with De Houwer et al. that the field
cott.
would profit from more clearly describing phenomena in terms
Jacoby, L. L. (1991). A process dissociation framework: Separating
of the environmental conditions under which they occur, cou- automatic from intentional uses of memory. Journal of Memory
pled with avoiding the use of theoretical terms in the naming of and Language, 30, 513–541.
phenomena. This sort of “call” is important to make every now Kingstone, A., Smilek, D., & Eastwood, J. D. (2008). Cognitive ethol-
and then, so that generations of researchers are aware of the pit- ogy: A new approach for studying human cognition. British Journal
falls. That said, we believe that the lion’s share of the research in of Psychology, 99, 317–340.
cognition already does this, possibly a legacy from our behav- Kornblum, S. (1994). The way irrelevant dimensions are processed
iorist roots such as the verbal learning tradition’s continuing depends on what they overlap with: The case of Stroop- and Simon-
influence on the cognitive study of memory. As such, we were like stimuli. Psychological Research, 56, 130–135.
not convinced of the need to apply the functional-cognitive dis- Lewin, K. (1951). Problems of research in social psychology. In D.
Cartwright (Ed.), Field theory in social science: Selected theoretical
tinction to all of applied psychology. On the positive side, how-
papers (pp. 155–169). New York: Harper & Row.
ever, we do see considerable value in papers of this sort for their
Lupiáñez, J., Klein, R. M., & Bartolomeo, P. (2006). Inhibition of return:
potential to make us think hard about what is important in our Twenty years after. Cognitive Neuropsychology, 23, 1003–1014.
discipline and to consider the meta-questions that we might oth- MacLeod, C. M., Dodd, M. D., Sheard, E. D., Wilson, D. E., & Bibi,
erwise neglect. Debate about core principles is good for science. U. (2003). In opposition to inhibition. In B. H. Ross (Ed.), The
psychology of learning and motivation (Vol. 43) (pp. 163–214). San
Diego, CA: Academic Press.
Conflict of Interest Statement Martín-Arévalo, E., Kingstone, A., & Lupiáñez, J. (2013). Is “inhibition
of return” due to inhibition of the return of attention? Quarterly
The authors declare no conflict of interest.
Journal of Experimental Psychology, 66, 347–359.
Mayr, S., & Buchner, A. (2015). Negative priming as a memory phe-
nomenon. Journal of Psychology, 215, 35–51.
Author Contributions Mayer, R. E., & Moreno, R. (2003). Nine ways to reduce cogni-
Order of authorship is alphabetical; the two authors con- tive load in multimedia learning. Educational Psychologist, 38,
tributed equally to the writing of the manuscript. 43–52.
 RADICAL COGNITIVISM 26

Meiser, T. (2011). Much pain, little gain? Paradigm-specific models and Risko, E. F., Richardson, D. C., & Kingstone, A. (2016). Breaking the
methods in experimental psychology. Perspectives on Psychological fourth wall of cognitive science real-world social attention and the
Science, 6, 183–191. dual function of gaze. Current Directions in Psychological Science,
Miyake, A., Friedman, N. P., Emerson, M. J., Witzki, A. H., How- 25, 70–74.
erter, A., & Wager, T. D. (2000). The unity and eiversity of Simon, J. R., & Berbaum, K. (1990). Effect of conflicting cues on
executive functions and their contributions to complex “frontal information processing: The ‘Stroop effect’ vs. the ‘Simon effect’.
lobe” tasks: A latent variable Analysis. Cognitive Psychology, 41, Acta Psychologica, 73, 159–170.
49–100. Sternberg, S. (1969). The discovery of processing stages: Extensions
Neisser, U. (1978). Memory: What are the important questions? In M. of Donders’ method. Acta Psychologica, 30, 276–315.
M. Gruneberg, E. E. Morris, & R. N. Sykes (Eds.), Practical aspects Titchener, E. B. (1929). Systematic psychology: Prolegomena. New
of memory (pp. 3–24). San Diego, CA: Academic Press. York: The MacMillan Company.
Newell, A. (1973). You can’t play 20 questions with nature and win: Tipper, S. P. (2001). Does negative priming reflect inhibitory mech-
Projective comments on the papers of this symposium. In W. G. anisms? A review and integration of conflicting views. Quarterly
Chase (Ed.), Visual information processing (pp. 283–308). New Journal of Experimental Psychology, 54A, 321–343.
York: Academic Press. Tunnell, G. B. (1977). Three dimensions of naturalness: An expanded
Newell, A. (1994). Unified theories of cognition. Cambridge, MA: definition of field research. Psychological Bulletin, 84, 426–437.
Harvard University Press. Watson, J. B. (1913/1994). Psychology as the behaviorist views it.
Posner, M. I., & Cohen, Y. (1984). Components of visual orienting. In Psychological Review, 101, 248–253.
H. Bouma, & D. Bouwhuis (Eds.), Attention and performance (Vol. Wickens, C. D. (2008). Multiple resources and mental workload.
10) (pp. 531–556). Hillsdale, NJ: Erlbaum. Human Factors, 50, 449–455.
Posner, M. I., Rafal, R. D., Choate, L. S., & Vaughan, J. (1985).
Inhibition of return: Neural basis and function. Cognitive Neuropsy-
chology, 2, 211–228. Received 31 October 2016;
Proctor, R. W., & Urcuioli, P. J. (2016). Functional relations and cog- accepted 8 November 2016
nitive psychology: Lessons from human performance and animal
research. International Journal of Psychology, 51, 58–63.

View publication stats