yes no Was this document useful for you?
   Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Document related concepts

Clark L. Hull wikipedia, lookup

Observational learning wikipedia, lookup

Residential treatment center wikipedia, lookup

Parent management training wikipedia, lookup

Reinforcement wikipedia, lookup

Neuroeconomics wikipedia, lookup

Applied behavior analysis wikipedia, lookup

Adherence management coaching wikipedia, lookup

Behavioral economics wikipedia, lookup

Professional practice of behavior analysis wikipedia, lookup

Operant conditioning wikipedia, lookup

B. F. Skinner wikipedia, lookup

Behavior and Philosophy, 41, 33-59 (2013). ©2015 Cambridge Center for Behavioral Studies
Diego Zilio
UFES -Federal University of Espírito Santo
ABSTRACT: It is often said, especially in philosophy and the neuroscience literature, that
Skinner defended an anti-physiological position on the explanation of behavior. Aside from
this, behavior analysts who discuss the relation between behavior analysis and physiology
usually emphasize the independence of these two fields. Amid criticisms of Skinner’s
allegedly anti-physiological position and behavior analysts’ defense of their discipline as
an autonomous science, there is comparatively little discussion of Skinner’s positive views
on physiology. The goal of this paper is to present an analysis of these views, taking into
account Skinner’s writings from the 1930’s to the 1990’s. Among the topics to be
discussed are the definition of the object of study of physiology, its role in the explanation
of behavior, and the relation between behavior-analytic and physiological explanations. I
pay special attention to Skinner’s use of physiological hypotheses in developing his
theories of private events and perception, and I hope to counteract the ill-founded notion
that Skinner was always opposed to physiology in the study of behavior.
Key words: Skinner, behavior analysis, radical behaviorism, neuroscience, physiology,
explanation of behavior
Outside the behavior-analysis community, Skinner is known for his critical
positions regarding physiological explanations of behavior (e.g., Bradnan, 1982;
García-Hoz, 2004; Illard & Feldman, 2001; Kandel, 1976; Loucks, 1941;
Machamer, 2009; Panksepp, 1990; Razran, 1965; Staddon & Bueno, 1993). To this
day, the notion that Skinner defended an anti-physiological position continues to
be disseminated among neuroscientists and philosophers. For example, consider
this excerpt from a handbook on the philosophy of neuroscience:
Skinner (…) dismisses the use of physiological variables, because he believes
that they could add nothing of relevance to the science of behavior. All one
needed, for Skinner, were the environmental or stimulus conditions (…), the
behavioral response, and the contingencies or conditions that led to
reinforcement. (…) Clearly in this radical Skinnerian sense, no neuroscientist
could be a behaviorist, for they, by virtue of discipline, must discuss variables
that are internal to the organism and irrelevant to a science of behavior.
(Machamer, 2009, p. 168)
Meanwhile, behavior-analytic writings often emphasize Skinner’s arguments
for the autonomy of behavior science from physiology (e.g., Baer, 1996; Carvalho
Neto, 1999; Carvalho Neto & Tourinho, 1999; Greenberg, 1983; Greenberg &
Lambdin, 2007; Reese, 1996; Staddon, 2001; Starling, 2000; Tourinho, Teixeira &
Maciel, 2000).
Amid criticisms of Skinner’s supposedly anti-physiological stance and
behavior analysts’ own efforts at establishing behavior analysis as an autonomous
science, Skinner’s positive views on the role of physiology in the explanation of
behavior have been rather neglected. This situation perpetuates the misguided
idea—evident in Machamer’s (2009) quote—that behavior analysis and
neuroscience are actually incompatible with each other. To be sure, a significant
literature already deals with the relation between behavior analysis and
neuroscience by clarifying their possible relations and roles in the explanation of
behavior (e.g., Bullock, 1996; Carvalho Neto, 1999; Carvalho Neto & Tourinho,
1999; Catania, 2000; Donahoe, 1996, 2002; Donahoe & Palmer, 1994; Elcoro,
2008; Mechner, 2008; Moore, 2002; Mustaca, 2003; Schaal, 2003, 2005; Silva,
1987; Strumwasser, 1994; Thompson, 2007, 2008, Timberlake, Schaal &
Steinmetz, 2005) and the present article can be seen as an addition to these efforts.
What is lacking so far is a detailed, systematic analysis of Skinner’s views on the
topic that takes into account his writings from the 1930’s to the 1990’s. He wrote
in several occasions that behavior analysis had “gaps” and that those would be
filled by physiology, especially neuroscience. The goal of this paper is to discuss
these “gaps”.
Where is the Boundary?
To understand Skinner’s perspective on the role of neuroscience in the
explanation of behavior, we first need to examine how he defined the object of
physiological studies and how it differs from that of behavior analysis. Through
various definitions, Skinner actually emphasized different aspects of the research
topics of behavior analysis and physiology. I will classify these definitions
respectively as structural, causal, and explanatory.
The structural definition
The structural definition presents the object of study of physiology as the
physiological structures that make behavior possible (Skinner, 1947/1961a,
1957/1961d, 1963, 1974, 1975a, 1983, 1988, 1989d). For example, Skinner wrote
that “as the science of physiology advances, it will presumably be possible to show
what is happening in various structures within the organism during particular
behavioral events” (1947/1961a, p. 236); “an organism behaves as it does because
of its current structure” (1974, p. 17); “it is a given organism at a given moment
that behaves, and it behaves because of its ‘biological equipment’ at that moment.
Eventually neurology will tell us all we need to know about the equipment” (1988,
p. 301); “what they call mechanism seems to be the current state of the organism
described in the language of physics and biology. The organism behaves as it does
because of the present state” (1988, pp. 304-305). According to Skinner (1974),
this structure is made of “effectors and receptors, nerves, and a brain” (p. 136).
Thus, a biological system (Skinner, 1956/1961e, p. 215)—the organism—is
necessary for behavior to occur. “Structure,” “equipment,” “state,” and
“mechanism” are terms used by Skinner to generically indicate the physiological
configuration of the organism while it behaves in a given fashion. Importantly, (a)
this configuration is causally responsible for the organism’s behavior, and (b) this
structure should not be view as static or permanent. It changes through time,
mostly as a consequence of the interactions between organism and environment.
Those changes in physiology are directly responsible for later behavioral changes.
Skinner made this point clear in numerous occasions (e.g., 1972, 1974, 1975b,
1985, 1988, 1988/1989c, 1993). For example:
A behavioral analysis acknowledges the importance of physiological research.
What an organism does will eventually be seen to be due to what it is, at the
moment it behaves, and the physiologist will someday give us all the details. He
will also tell us how it has arrived at that condition as a result of its previous
exposure to the environment as a member of the species and as an individual.
(Skinner, 1974, p. 249)
The behavior of an organism at time t can be causally associated with what
the organism is at t. Here no mention is made of phylogenetic or ontogenetic
history. Explaining behavior at moment t is done solely in terms of the organism’s
physiological “state” or “configuration” at t. Skinner (1988) goes even further
along these lines:
Of course differences in structure are important, but they are themselves the
results of contingencies of selection. Structures are what is selected. Organisms
have evolved in such a way that when they are hungry, food strongly reinforces
their behavior. Other reinforcers are merely conditioned. But both reinforce
because of the structure of the organism. The structure is not an initiating cause
of anything, however. (p. 434, italics added)
It is the organism’s physiological configuration, or what Skinner calls
“structure,” that is the object of selection by phylogenetic and ontogenetic
contingencies. This object is not behavior, at least not directly. Contingencies
select behavior only to the extent that they select physiological structures or
“configurations” responsible for behavior (cf. Glenn & Madden, 1995). The point
here is that in Skinner’s view the selection of behavior is indirect; it depends on the
intermediary link, the physiological configuration of the organism. This does not
imply, however, that all of the causal factors related to behavior are to be found in
physiology. We cannot say of the physiological configuration at time t that it is the
initial and only cause of behavior at t, because this configuration is itself a product
of phylogenetic and ontogenetic contingencies.
The causal definition
According to Skinner (1986), “in a given episode the environment acts upon
the organism, something happens inside, the organism then acts upon the
environment, and certain consequences follow” (p. 716). This excerpt suggests a
causal chain made of environmental events, physiological processes and behavior.
Environmental events change the physiological configuration of an organism;
these changes cause some behavior; the behavior, by its turn, changes the
environment. Figure 1 illustrates this causal chain.
Figure 1: Skinner’s (1986) causal chain.
According to Skinner (1986), “the first, third, and fourth of these events is the
field of a science of behavior, which undertakes to discover how they are related to
each other. What happens inside is another part of the story” (p. 716). In terms of
Figure 1, behavior analysis deals with the events labeled (1), (3), and (4); events
which, by definition, constitute the three-term contingency, the basic unit of
analysis of operant behavior (cf. Skinner, 1969a). The remaining link in the causal
chain, event (2), is the proper domain of physiology, especially neuroscience.
When Skinner (1988) says that “the nervous system mediates the relations studied
in an experimental analysis of behavior” (p. 257), he is fully consistent with Figure
Whereas Figure 1 presents events rather simplistically as a single chain-like
sequence, Skinner (1956/1961e, p. 206) formulates a richer and more heuristic
description of the same basic idea. Here it is, using Skinner’s own illustration this
time (Figure 2).
Figure 2: Skinner’s heuristic causal chain. Adapted from Skinner (1956/1961e, p. 206).
Instead of a point-to-point description of the events that constitute the causal
chain, here we find a distinction between the behavior of organism (the dependent
variable) and the environmental events related to this organism’s phylogenetic and
ontogenetic histories (the independent variables). According to Figure 2, the
behavioral repertoire of an organism at time t is a product of its species’ history
and of its own history of interactions with the environment. But what about the
“organism”, the mediational variable of this causal chain? In Skinner’s
(1956/1961e) words:
This does not mean, of course, that the organism is conceived of as actually
empty, or that continuity between input and output will not eventually be
established. The genetic development of the organism and the complex
interchanges between organism and environment are the subject matters of
appropriate disciplines. Some day we shall know, for example, what happens
when a stimulus impinges upon the surface of an organism, and what happens
inside the organism after that, in a series of stages the last of which is the point
at which the organism acts upon the environment and possibly changes it. At
that point we lose interest in this causal chain. Some day, too, we shall know
how the ingestion of food sets up a series of events, the last of which to engage
our attention is a reduction in the probability of all behavior previously
reinforced with similar food. Some day we may even know how to bridge the
gap between the behavioral characteristics common to parents and offspring.
But all these inner events will be accounted for with techniques of observation
and measurement appropriate to the physiology of the various parts of the
organism, and the account will be expressed in terms appropriate to that subject
matter. (…) Its task [physiology’s] is to account for the causal relations between
input and output which are the special concern of a science of behavior.
Physiology should be left free to do this in its own way. (pp. 214-215)
One day it will be possible to know how the environment affects the organism
by causing physiological changes, which are in turn responsible for the occurrence
of behavior. Skinner also brings into discussion the causal/mediational role of
physiological events: They are paramount to the very existence of causal relations
between the independent and dependent variables (what Skinner calls “inputs” and
“outputs” in this excerpt) studied in behavior analysis.
The explanatory definition
Skinner often discussed the possibility of direct manipulation of physiological
variables with the aim of predicting and controlling behavior. His discussions
exemplify the third aspect of what would be the object of study of physiology from
Skinner’s perspective. Let us start with the following excerpt:
Eventually a science of the nervous system based upon direct observation rather
than inference will describe the neural states and events, which immediately
precede instances of behavior. We shall know the precise neurological
conditions which immediately precede, say, the response, “No, thank you.” (…)
However, we may note here that we do not have and may never have this sort of
neurological information at the moment it is needed in order to predict a specific
instance of behavior. It is even more unlikely that we shall be able to alter the
nervous system directly in order to set up the antecedent conditions of a
particular instance. The causes to be sought in the nervous system are, therefore,
of limited usefulness in the prediction and control of specific behavior. (Skinner,
1953/1965, pp. 28-29, italics added)
Here we find a more detailed description of the idea already present in
Skinner (1956/1961e) about the possibility of knowing one day the physiological
events causally responsible for a particular behavior—the events labeled (2) in the
causal chain of Figure 1. Skinner talks specifically about neuroscience and, most
importantly, he says that neurophysiological events must be directly observed. It is
impossible to predict and control behavior by manipulating neurophysiological
events without any direct knowledge of neurophysiology. Skinner (1953/1965)
goes further:
To what extent is it helpful to be told, “He drinks because he is thirsty”? If to be
thirsty means nothing more than to have a tendency to drink, this is mere
redundancy. If it means that he drinks because of a state of thirst, an inner causal
event is invoked. If this state is purely inferential - if no dimensions are assigned
to it which would make direct observation possible - it cannot serve as an
explanation. But if it has physiological (…) properties, what role can it play in a
science of behavior? The physiologist may point out that several ways of raising
the probability of drinking have a common effect: they increase the
concentration of solutions in the body. Through some mechanism not yet well
understood, this may bring about a corresponding change in the nervous system
which in turn makes drinking more probable. (…) we have a causal chain
consisting of three links: (1) an operation performed upon the organism from
without— for example, water deprivation; (2) an inner condition—for example,
physiological (…) thirst; and (3) a kind of behavior—for example, drinking.
Independent information about the second link would obviously permit us to
predict the third without recourse to the first. (…) The second link is useless in
the control of behavior unless we can manipulate it. (pp. 33-34, italics added)
Skinner again describes the heuristic causal chain discussed earlier, but now
he focuses on the manipulation of links in the chain. An aspect of the organism’s
history (in this case, the operation of deprivation, 1) modifies in some way the
neurophysiology of the organism (2), resulting in what we may call “state of
thirst”, which in turn increases the probability of drinking behavior (3). The second
link in the explanation of behavior can be dealt with in different ways, however.
We cannot infer the “state of thirst” from the frequency of drinking behavior. This
would be circular reasoning: inferring that the organism is thirsty because it is
drinking water, and explaining the drinking by saying that the organism is thirsty.
Inferring the second link from the first (for example, to say that the organism is
“thirsty” because of the long time since it last drank water) would be useless as
well, because this inference would depend on our previous knowledge of the
history of organism. As far as Skinner is concerned, an inferential second link
cannot serve in the explanation of behavior: To be useful, explanations in terms of
the second link should deal with events that are observable and susceptible of
manipulation. Otherwise, we face the risks of explanations based on hypothetical
constructs (Moore, 2008; Skinner, 1974, 1950, 1953/1965, 1988; Zuriff, 1985):
“The most objectionable practice is to follow the causal sequence back only as far
as a hypothetical second link” (Skinner, 1953/1965, p. 34).
What if the second link is actually physiological? What if the “state of thirst”
is described accurately as a certain physiological process? In that case, according
to Skinner, we could associate different ways of raising the probability of drinking
behavior (deprivation of water, ingestion of salt, etc.) with their physiological
effects and associate drinking behavior with the physiological events that precede
it. When Skinner wrote his comments, the relevant knowledge was unavailable, but
he was aware that this was only a technological limitation of the physiology of the
times (Skinner, 1969b, 1972, 1974, 1975b, 1987, 1988). Eventually, directly
knowledge of the second link would be possible and then (only then) it could serve
to the explanation of behavior. This point is clear in Skinner’s (1972, 1974) talk of
the “omniscient physiologist”:
There are gaps in time and space between behavior and the environmental
events to which it is attributed, and it is natural to try to fill them with an
account of the intervening state of the organism. We do this when we summarize
a long evolutionary history by speaking of genetic endowment. Should we not
do the same for a personal history? An omniscient physiologist should be able to
tell us, for example, how a person is changed when a bit of his behavior is
reinforced, and what he thus becomes should explain why he subsequently
behaves in a different way. (…) An omniscient physiologist should be able to do
the same for comparable states in the field of behavior. He should also be able to
change behavior by changing the organism directly rather then by changing the
environment. (1972, p. 18)
The omniscient physiologist is Skinner’s argumentative device to defend the
possibility, at least in principle, of explaining behavior by studying the second link.
The omniscient physiologist is able to establish causal relations between events
related to the history of organism and changes in his physiological configuration
(first and second links) as well as between those physiological events and the
organism’s subsequent behavior (second and third links). Therefore, the
omniscient physiologist can predict and control behavior through the observation
and manipulation of physiological events.
Explanations and Gaps
Skinner’s most common attribution of function to neuroscience is that of
filling in the gaps of a behavioral analysis (1953/1965, p. 54; 1956/1961e, p. 214;
1969b, p. 24; 1972, p. 18; 1974, pp. 214-215; 1975b, p. 43; 1985, p. 297; 1987, p.
782; 1988, p. 122, 128, 257, 470; 1989a, p. 56; 1989d, p. 18; 2009, p. 69). For
The promise of physiology is of a different sort. New instruments and methods
will continue to be devised, and we shall eventually know much more about the
kinds of physiological processes, chemical or electrical, which take place when
a person behaves. The physiologist of the future will tell us all that can be known
about what is happening inside the behaving organism. His account will be an
important advance over a behavioral analysis, because the latter is necessarily
“historical”—that is to say, it is confined to functional relations showing
temporal gaps. Something is done today which affects the behavior of an
organism tomorrow. No matter how clearly that fact can be established, a step is
missing, and we must wait for the physiologist to supply it. He will be able to
show how an organism is changed when exposed to contingencies of
reinforcement and why the changed organism then behaves in a different way,
possibly at a much later date. What he discovers cannot invalidate the laws of a
science of behavior, but it will make the picture of human action more nearly
complete. (Skinner, 1974, pp. 214-215, italics added)
Gaps arise because behavior analysis leaves out the second link of the causal
chain in Figure 2. Basically, behavior analysts manipulate contingencies (the first
link in the chain) in order to the see the effects on behavior (third link). No matter
how detailed, a functional analysis of the relations between the first and third links
will always be incomplete, because it does not include information about the
second, physiological, link. Notice that the gap in analysis is tolerable only if we
assume transitivity between the causal links in the chain. In this context,
transitivity means that if A causes B and B causes C, then A causes C. If
contingencies of selection (A) are responsible for changes in an organism’s
physiology (B) and if the latter changes result in behavior (C), then, assuming
these events are orderly related to each other, contingencies of selection are
ultimately responsible for the behavior of organisms. “Unless there is a weak spot
in our causal chain so that the second link is not lawfully determined by the first,
or the third by the second, then the first and third links must be lawfully related”
(Skinner, 1953/1965, p. 35).
Behavior analysis qualifies as an incomplete science to the extent that it lacks
information about the physiological variables responsible for behavior. But
behavior analysis is still capable of providing “explanations” by studying the
relations between contingencies and behavior change. The explanatory gap will be
filled by physiology. So in this sense, physiology can also provide “explanations”
of behavior. Here, however, it becomes paramount to discuss how behavioranalytic and physiological explanations relate to each other. In particular, are we
talking about the same notion of “explanation”?
From a radical behaviorist perspective, scientific explanation basically
consists in describing functional relations between events (Skinner, 1931/1961c,
1938/1966a, 1947/1961a, 1953/1965, 1966b; cf. Donahoe, 2004; Hayes &
Brownstein, 1986; Lacey, 1984; Laurenti & Lopes, 2008; Moore, 2008), and an
explanation is valued mainly to the extent that it promotes prediction and control
of the phenomena being explained (Skinner, 1938/1966a, 1953/1965; cf. Hayes &
Brownstein, 1986; Lacey, 1984; Moore, 2010; Palmer, 2009; Smith, 1992). But
caution is needed when dealing with the to-explain-is-to-describe motto often
found in the behavior-analysis literature (e.g., Baum, 1994; Catania, 2007; Reese,
1996), because the mere description of relations between events does not
invariably count as an explanation of these events (cf. Marr, 2003; Tonneau, 2008).
Let us say that a scientist observes the occurrence of an event A and,
following it, the occurrence of another event B. After a long period of observation,
the scientist notes that A always occurs before B, that B always follows A, and that
A and B never occur alone. These observations eventually start to control the
scientist’s “explanatory verbal behavior”. She starts saying, “every time A occurs,
B occurs” and, “B is always preceded by A”. Explaining takes more than mere
description, however; explaining demands an active role of the scientist in
manipulating the events under study (Woodward, 2003). We can assume a
functional relation between A and B only if we see effects in B when manipulating
A. Even prediction is not enough for explanation. Through a series of observations,
a scientist can “predict” the occurrence of B after seeing A occur. It is thus
possible to predict events from observation alone, without any kind of
manipulation. A consistent but fortuitous correlation between events can provide
useful information for strictly predictive purposes. Explanation, however, goes
beyond description and prediction to include manipulation and control. If the cooccurrence of A and B is merely fortuitous, then the manipulation of A will have
no impact on the occurrence of B; but if it turns out that we can control B by
manipulating A, then we have solid grounds to assume that A and B are
functionally related.
In sum, explanation is not merely a narrative activity. Explanation requires the
discovery of conditions (or variables) necessary for the production of a given
phenomenon. For Skinner (1957), to explain is to specify “what conditions are
relevant to the occurrence of the behavior—what are the variables of which it is a
function?” (p. 10). The process of “discovering” these conditions in turn requires
manipulating events and analyzing the effects of those manipulations. To explain a
phenomenon B, as a form of verbal behavior, thus consists in describing the
variables responsible for the production of B. The scientist in our example would
say, “every time I produce A, B also occurs” or, “every time I prevent A, B also
fails to occur”. Achieving the stage of functional description (that is, explanation)
requires the active role of the scientist, who is not merely a passive observer.
Assuming this view of explanation, to say that the behavior-analytic
explanation of behavior has gaps means that it does not include some of the
variables that are responsible for the production of behavior. These variables (and
among them, those studied by physiology or neuroscience) should be seen as
important as the ones studied in behavior analysis, for the good reason that in their
absence there would be no behavior to explain. Alongside behavior analysis,
therefore, physiology can provide explanations of behavior by discovering some of
the variables responsible for its occurrence. Physiology discovers functional
relations between contingencies of selection and physiological processes (first and
second links) as well as between physiological processes and behavior (second and
third links).
Are physiological explanations reductive? Skinner (1963) wrote, “The
physiologist studies structures and processes without which behavior could not
occur. He is in a position to supply a ‘reductionist’ explanation beyond the reach
of an analysis which confines itself to terminal variable” (p. 957, italics added).
Skinner’ use of the term reduction may be misguided, however. Analyzing the
concept of reduction is no easy task (e.g., Bickle, 1998; Horst, 2007; Schaffner,
1967), but it is hard to see how an explanation of behavior that focuses on
contingencies (or “history”) could reduce to an explanation based on neural
processes. We cannot reduce the occurrence of a stimulus (a historical event) to
any given brain activity, for example (Marr & Zilio, 2013). Admittedly, in some
cases physiological explanations might be used instead of behavior-analytical
explanations, particularly when we are ignorant of historical variables. Skinner
(1953/1965) even said that physiological explanations would be preferable in this
Independent information about the second link would obviously permit us to
predict the third without recourse to the first. It would be a preferred type of
variable because it would be non-historic; the first link may lie in the past
history of the organism, but the second is a current condition”. (p. 34, italics
In the absence of knowledge about the real nervous system, however, an
explanation based on historical variables is better than those appealing to an
inferential, conceptual nervous system.
Demarcation Criteria
When Skinner (1986) delineated the causal chain between environmental,
physiological and behavioral events, he also asserted “behavior and physiology are
not two ways of approaching the same subject” (p. 716, italics added). The same
idea can be found in a later excerpt: “brain processes are not another 'aspect’ of
behavior; they are another part of what an organism does” (Skinner, 1989a, p. 56).
Here Skinner suggests a clear-cut separation between the events studied by
behavior analysis (i.e., environmental and behavioral events) and those studied by
physiology; these two sciences do not deal with the same object and therefore
cannot be distinct ways of approaching the same phenomenon.
At other places, however, Skinner suggests that physiological events are part
of behavior and not of a different domain: “I have never questioned the importance
of physiology or in particular brain science or its relevance to behavior. What is
happening inside the skin of an organism is part of its behavior” (Skinner, 1989a,
pp. 129-130, italics added). This quote suggests the opposite of the earlier one
(1989a, p. 56). Are we facing a contradiction in behavior-analytic thinking? I will
try to answer this question by discussing two kinds of empirical work, one that
addresses physiological events as parts of behavioral contingencies, the other
dealing with physiology as a mechanism underlying behavior.
With respect to empirical work of the first kind, Skinner (1953/1965) defined
the environment as “any event in the universe capable of affecting the organism”
(p. 257). The effective environment thus comprises conditional and discriminative
stimuli, and this sense of “environment,” operant contingencies can incorporate
physiological events as antecedent stimuli, consequences, or responses (BarnesHoles, 2003; Branch, 2006; Silva, Gonçalves & Garcia-Mijares, 2007). The effects
of substance administration, for example, can function as discriminative stimuli for
operant behavior, an important issue not only with respect to drug taking but also
with respect to private events (Branch, 2006). Lubinski and Thompson’s (1987,
1993) experimental model for the interpersonal communication of private states
thus involves (among other things) the discriminative control of behavior by the
physiological effects of pentobarbital, cocaine, and saline administration. The
important point here is that physiological effects can function as discriminative
stimuli; they can be part of the behavioral relations studied in behavior analysis.
Notice that Lubinski and Thompson’s goal was purely behavioral: “[to] determine
whether nonhuman animals could learn to interact communicatively (i.e., exchange
discriminative stimuli with each other), based on events in their internal milieu”
(1987, p. 1). The background knowledge that informed Lubinski and Thompson’s
empirical demonstration was also behavioral: from the definition of private events,
tact and mand, to research on symbolic communication in non-human animals and
to the discriminative effects of drug administration (Lubinski & Thompson, 1987,
As an example of empirical work of the second kind, consider an experiment
by Campeau, Miserendino and Davis (1992) that also deals with substance
administration, but differs from Lubinski and Thompson’s (1987) study in the way
physiological variables are addressed. The difference is already evident in the
authors’ stated purpose: “to test the effects of intra-amygdala infusion of AP5 on
the acquisition and expression of aversive conditioning to an auditory stimulus as
measured with fear-potentiated startle” (Campeau et a1. 1992, p. 569). In this case,
the administration of a chemical substance (AP5) is justified by its effects on a
particular neurophysiological mechanism and not because of its function as part of
behavioral contingencies. Indeed, it is known that the amygdala—a group of nuclei
located within the medial temporal lobe—is part of the mechanism for establishing
the behavioral relations associated with “fear” (LeDoux, 2000). The procedure
(called “fear conditioning”) used to document this role consists of pairing an
aversive stimulus, usually shock, with a neutral stimulus (LeDoux, 2000) and
recording the resulting changes in heart rate, arterial pressure, skin conductance,
hormonal secretion, and motor activity (Davis & Whalen, 2001; LeDoux, 2000).
A detailed description of the physiological mechanisms of “fear conditioning”
would be beyond the scope of this article. Suffice it to say that in order to increase
the synaptic efficacy relevant to fear conditioning, NMDA receptors in the lateral
amygdala must be unblocked, since they are usually blocked by Mg2+ molecules
(Sah, Westbrook, & Lüthi, 2008). Conversely, any manipulation that maintains the
NMDA receptors in a blocked state should make fear conditioning impossible.
That is exactly the hypothesis tested by Campeau et al. (1992). They maintained
the NMDA receptors in a state of blocking by the intra-amygdala infusion of AP5
((2R)-amino-5-phosphonovaleric acid), a NDMA receptor antagonist. The results
showed that the infusion of AP5 before the “learning period” (when the subjects
were submitted to the aforementioned contingency) “blocked” the establishment of
“fear conditioning.”
Based on these two examples (many more could be found in the technical
literature), we can see that (a) studying the operant discrimination of effects
produced by the administration of certain substances and (b) studying the effects of
the administration of certain substances on a neurophysiological mechanism are
not quite the same thing. We may allude to this distinction as involving a double
aspect of physiological events, provided we remember that the two “aspects” are
not intrinsic to the events themselves, but only concerns how two scientific
communities interact with a shared object of study (Zilio, 2013). Thus, the
distinction ultimately involves the different contingencies that control the scientific
behavior of behavior analysts and neuroscientists. Physiological events can be
involved in behavioral contingencies, as stimuli (antecedent and consequent) or
responses. If studied as such, they are part of the domain of behavior analysis; they
are “part of behavior,” as Skinner put it. If, however, physiological events are
studied as mechanisms (or components thereof) associated with the occurrence of
behavior, then they qualify as physiology, in particular neurophysiology. Now they
fit Skinner’s definition of physiological events as intermediary links between
historical variables and the behavior of an organism. In this context, Skinner’s
“causal chain” is best viewed not as a representation of distinct events linearly
related to each other, but as an abstraction that distinguishes the objects of study of
behavior analysis and physiology according to how two scientific communities
interact with the same objects.
The distinction is clearer when we shift from a discussion of the “object” of
analysis to a discussion of the “role” of physiology in the explanation of behavior.
According to Skinner, the overall goal of physiology as a scientific discipline is to
explain the physiological mechanisms associated with the occurrence of behavioral
relations: “How contingencies of operant reinforcement affect physiological
processes is no doubt an important question” (Skinner, 1988/1989c, p. 82). Again:
The behaving organism will eventually be described and explained by the
anatomist and physiologist. As far as behavior is concerned, they will give us an
account of the genetic endowment of the species and tell how that endowment
changes during the lifetime of the individual and why, as a result, the individual
then responds in a given way on a given occasion. (Skinner, 1975b, p. 42)
This excerpt is important because it describes the “how” and “why” of
physiology according to Skinner. The “how” points to the study of physiological
mechanisms that allow an organism to respond to the environment and how these
mechanisms change during the organism’s interactions with the environment. The
“why”, on the other hand, stands for explaining why an organism behaves in such a
way at a given time. “Why” questions are typically associated with the behavioranalytic study of phylogenetic and ontogenetic contingencies. Skinner, however,
implies that physiology can also offer its own answers to “why” questions by
clarifying the functional relations between physiological processes and behavior.
The same idea showed up elsewhere in the Skinnerian corpus: “A great deal
goes on inside the skin, and physiology will eventually tell us more about it. It will
explain why behavior is indeed related to the antecedent events of which it can be
shown to be a function” (Skinner, 1971b, p. 195). Put differently, physiology will
explain “why” there are ordered functional relations between the first and third
links (phylogenetic and ontogenetic histories, and behavior, respectively) in
Skinner’s causal chain.
Skinner’s Reliance on Physiology: Some Examples
Beyond his general comments on the objects and goals of physiology and
behavior analysis, Skinner also raised specific issues about behavior that may have
required a physiological answer. Some of his interpretations and conceptual
analyses about behavior also involved physiological hypotheses. In this section I
will discuss a few cases of Skinner’s interpretations that rely on physiology.
Motivation, discrimination and reinforcement
One of the issues that Skinner thought would be answered by physiology
concerned the role of “motivational” variables in the explanation of behavior:
One of my first papers [Skinner, 1932] was on the state of hunger (or ‘drive'),
and I have been interested in states off and on ever since. The organism we
observe and possibly study can certainly be said to be in a given state at a given
time. The state will eventually be directly observed by those who have the
proper instruments and methods - namely, anatomists and physiologists. They
will complete the account offered by an experimental analysis of behavior,
which necessarily has temporal and spatial gaps. (Skinner, 1988, p. 122)
“Hunger” or “drive” are commonly viewed as motivational states. What
Skinner is saying here is that these states should be studied directly (through direct
observation and manipulation instead of inference) and therefore physiologically. I
have already discussed this example in relation to a history of deprivation and the
resulting behavior. Elsewhere, Skinner (1966b) explicitly relates the issue to
motivation and emotion: “other independent variables are found in the classical
fields of motivation and emotion. The experimental analyst does not manipulate
inner states as such” (p. 215).
A similar idea can be found in Skinner’s (1988) remarks on a study of
discriminative control:
The experiment throws additional light on the nature of the control exerted by
stimuli. Eventually a physiological explanation will be available. Meanwhile, I
do not see that anything is gained by referring to a hidden mechanism. On the
contrary, representing a datum with a hidden mechanism makes it harder to
integrate with other data from other experiments, which could also be done
without postulating a hidden mechanism (p. 133).
Later on Skinner (1988) argues: “as a behavioral problem, discrimination
raises no question if one stops talking about input-processing mechanisms. The
behavioral facts are well known. They are often astonishing, and a neurological
explanation is badly needed” (p. 471). According to Skinner, physiology could
also explain (at least in part) the difference between elicitation and discrimination:
“a discriminative stimulus which brings about the emission of a response (which
'sets the occasion' for the response) differs quantitatively in its action from the
eliciting stimulus and must be 'explained' by a different neural mechanism”
(Skinner, 1938/1966a, p. 430). In this context, Skinner (1988) again stresses that
only the physiological variables that are directly observable and susceptible of
manipulation (as opposed to hypothetical, inferential links) are worth pursuing. No
less than in the case of “motivational” and “emotional” states, a non-hypothetical
physiology should be able to shed light on the mechanisms that mediate behavioral
processes of elicitation and discrimination.
The study of the physiological mechanisms of reinforcement is another clear
example of how Skinner relied on neuroscience to elucidate important behavioral
issues. This seems to especially important in the light of the alleged circularity in
the concept of reinforcement (e.g., Skinner, 1953/1965, 1957/1961d, 1971a, 1972,
1973, 1988; see also: Paniagua, 1985; Prado Jr., 1982; Schick, 1971; Tonneau,
2008). In Skinner’s (1953/1965) words:
The only defining characteristic of a reinforcing stimulus is that it reinforces.
The only way to tell whether or not a given event is reinforcing to a given
organism under given conditions is to make a direct test. We observe the
frequency of a selected response, then make an event contingent upon it and
observe any change in frequency. If there is a change, we classify the event as
reinforcing to the organism under the existing conditions. There is nothing
circular about classifying events in terms of their effects; the criterion is both
empirical and objective. It would be circular, however, if we then went on to
assert that a given event strengthens an operant because it is reinforcing. (pp.
Behavior analysis provides a classificatory description of presently
reinforcing events. If we want to go further and know why reinforcers reinforce,
then we should turn to other scientific fields, including physiology: “…a biological
explanation of reinforcing power is perhaps as far as we can go in saying why an
event is reinforcing” (Skinner, 1953/1965, p. 84); “…in the field of behavior, it is
the reinforcing consequences which alter the probability that an organism will
subsequently behave in a given way. The physiological mechanisms which make
this possible (themselves a product of natural selection) are not yet known”
(Skinner, 1973, p. 264); finally:
Another possibility is that neurological conditions may be arranged which will
have a positive effect. A step in this direction has been taken by James Olds with
his discovery that weak electrical stimulation of certain parts of the brain,
through permanently implanted electrodes, has an effect similar to that of
positive reinforcement. In one of Olds' experiments, a rat presses a lever to give
itself mild electrical stimulation in the anterior hypothalamus. When every
response is so "reinforced," behavior is sustained in strength for long periods of
time. (…) Although there remain some puzzling differences between behavior
so reinforced and behavior reinforced with food, Olds' discovery in an important
step toward our understanding of the physiological mechanisms involved in the
operation of the environmental variable. (Skinner, 1957/1961d, pp. 122-124)
Skinner rarely mentioned specific examples of research in neuroscience, so
his reference to Olds’ work should not be taken lightly. Skinner may have given
credit to Olds’ research because the latter was dealing with the nervous system in
actual physiological terms and not as a conceptual/inferential system.
Private events
In Skinner’s remarks on private events we find him not only raising issues to
be addressed by physiology but also advancing physiological hypotheses that
would support his own conceptions. In a generic sense, private events involve
behavioral relations (i.e., relations between stimuli and responses) that are
“accessible” only to the behaving person (cf. Moore, 1980, 2009; Palmer, 2009;
Tourinho, 2006a, 2006b). I have argued elsewhere (Zilio, 2010; Zilio & Dittrich,
2014) that the core of Skinner’s theory of private events is not so much their lack
of accessibility than the special kind of “contact” that makes private relations
possible. Here is Skinner’s (1963) take on the subject:
The fact of privacy cannot, of course, be questioned. Each person is in special
contact with a small part of the universe enclosed within his own skin. To take a
noncontroversial example, he is uniquely subject to certain kinds of
proprioceptive and interoceptive stimulation. Though two people may in some
sense be said to see the same light or hear the same sound, they cannot feel the
same distension of a bile duct or the same bruised muscle. (When privacy is
invaded with scientific instruments, the form of stimulation is changed; the
scales read by the scientist are not the private events themselves). (p. 952, italics
Behavioral relations do not float around but require anatomical and
physiological support, and for something in the world to become a “stimulus,”
some kind of sensory apparatus is needed. According to Skinner (1953/1965, 1963,
1969b, 1972, 1974), there are three ways of “contacting” the world: (a) via the
exteroceptive nervous system, responsible for contacting the environment that is
also accessible to others (through their own exteroceptive nervous system); (b) via
the interoceptive nervous system, responsible for contacting the digestive,
circulatory and respiratory systems; and (c) via the proprioceptive nervous system,
responsible for contacting posture and movement. It is proprioceptive and
interoceptive stimulations that make private behavioral relations possible, and
these relations are “private” to the extent that they are supported by a specific
physiological apparatus. “Proprioceptive and interoceptive stimuli may have a
certain intimacy. They are likely to be especially familiar. (…) They may well be
of a special kind” (Skinner, 1963, p. 953). In Skinner’s (1963) example, two
persons may be said to see the same light because both are contacting the stimulus
via their exteroceptive nervous system. One cannot feel another person’s “pain,”
however, because the contact with the source of stimulation differs among people.
Dentists access the inflamed teeth of their patients via the dentists’ exteroceptive
nervous systems. In contrast, patients access their own inflamed teeth via their own
interoceptive and proprioceptive nervous systems (unless the patients also happen
to look at their teeth in a mirror).
One important consequence of the “special contact” definition of private
events is that they are not accessible through scientific instruments such as
microscopes, brain scans, and so on. The latter only enhance scientists’
exteroceptive contact with the source of stimulation. They do not turn private
relations into public ones. Regardless of which apparatuses she relies on, the
dentist does not contact the source of stimulation—an inflamed tooth—the way her
patient does. As Skinner (1957/1961b) puts it: “As physical states in the
individual, [private events] are a part of the physical world, but the individual
himself has a special connection with them. My aching tooth is mine in a very real
sense because none of you can possibly get nerves into it” (p. 201, italics added).
Note that “private” does not means “internal” or “inside the skin”. These
connotations are metaphorical (and imprecise) descriptions at best. One stimulus is
private, not because it is part of an organism’s physiology as opposed to the
external environment, but because the organism has as “special contact” with this
stimulus. This is why it is impossible to “invade privacy.” As Skinner (1963)
argued, as soon as others access the source of stimulation, the form of stimulation
is changed—it is not private anymore. None of this implies, however, that
exteroceptive studies of a source of private stimulation are pointless; on the
contrary, these studies can supply important information about private events
(Skinner, 1957). The dentist can diminish the patient’s “pain” by delivering an
anesthetic to the inflamed tooth. Merely seeing the inflamed tooth is additional
evidence (along with vocal and facial expressions) that the patient is “telling the
truth,” that is, that he is “really in pain”. So, it usually matters to know the
characteristics of a source S of private stimulation because S is an additional
variable that can control our evaluating others’ verbal descriptions of private
events. Knowing more of S is also important for teaching members of our social
community to discriminate private stimuli more precisely: “if events hitherto
classified as private can now be directly observed by the verbal community, the
community can arrange better contingencies in teaching its members to talk about
them” (Skinner, 1969a, p. 262).
A similar analysis applies to covert behavior, with some differences worth
discussing. According to Skinner (1953/1965, 1963, 1969a, 1972, 1974), we
contact physiological stimulation privately through our interoceptive and
proprioceptive nervous systems. Could there also be “pathways” responsible for
private relations and located wholly in the central nervous system (Moore, 2001,
2002, 2009; Natsoulas, 1983, 1985) Schnaitter (1984) answers affirmatively:
Skinner seems to assume that a sensory receptor in the peripheral nervous
system must be stimulated in order for a feeling to occur. This is simply not so.
For example, receptors exist within the central nervous system itself.
Osmoreceptors located in the hypothalamus are functionally related to drinking
and presumably to the experience of thirst. But it is not necessary for any
receptors to be stimulated at all for experiences to occur? The phenomenon of
dreaming is probably the clearest illustration of this. The current
neurophysiology of dreaming suggests that an endogenously cycling brainstem
structure regulates midline thalamic pacemaker projections to the cerebral
cortex: cortical desynchrony is correlated with the reported dream state. (p. 7)
The assumption that private relations involve only interoceptive and
proprioceptive stimulation is hardly essential to Skinner’s views on privacy,
however. Consider the following:
I agree that the kind of thinking which seems to be merely covert behavior
(“truncated, unemitted, reduced, impotent behavioral acts”) may be so reduced
that there is no muscular involvement to be sensed proprioceptively. Must we
appeal to some minute behavior which never reaches a muscle? If so, it is a
problem for the physiologist. There is a possibility that the effect is sensory. We
may see ourselves behaving rather than actually behave. I believe that my
analysis of seeing makes this a possible alternative. (Skinner, 1988, p. 331)
Thus, private behavioral relations can occur without any involvement of
muscular activity or even interoceptive and proprioceptive stimulation. Arguably,
this is the core of what we usually define as “covert behavior” (loosely speaking:
“thinking”, “problem solving”, “imagining”, “internal speech”, and so on). The
nervous system is massively interconnected. Brain areas responsible for a given
functions are connected to other areas responsible for other functions (cf. Baars &
Gage, 2010; Gazzaniga, Ivry, & Mangun, 2002). Brain areas related to “verbal
behavior,” for example, are connected to other areas associated with “mental
images” (cf. Bartolomeo, 2008; Farah, 2000; Gulyás, 2009; Kosslyn, Thompson &
Ganis, 2006). So when a person covertly says to himself (“internal speech”) that he
is seeing Picasso’s Guernica in the absence of the actual painting (“seeing in the
absence of the thing seen”), he may very well be responding discriminatively (and
privately) to the neurophysiological effects once elicited by the actual painting but
that are now under control of other variables.
To dispel possible misunderstanding, I have not been arguing that differences
between private and public events are only a matter of anatomy or physiology.
What I have been saying is that anatomo-physiological considerations are a key
factor in Skinner’s distinction between private and public events. Physiology plays
an important part in the study of private relations by providing knowledge about
variables that we can use in teaching and assessing descriptions of private events
(and not because it can be used to make private relations public; see Zilio, 2010;
Zilio & Dittrich, 2014).
Skinner’s theory of perception is an especially important example of his
appealing to physiology in solving theoretical issues in behavior analysis. His
theory essentially is a behavioral analysis of perception (cf. Lopes & Abib, 2002;
Zilio, 2010) that incorporates respondent and operant perceptual relations. In both
cases, the same logic applies: an antecedent condition (1) sets the occasion for a
perceptual response (2) that in turn may act as a discriminative stimulus (3) for a
person’s verbal description of perceptual experience (4).
Among the antecedent conditions of perception are unconditional stimuli. In
this case, stimuli are responsible for eliciting unconditional perceptual responses
(Skinner, 1953/1965). The visitor of the Reina Sofía Museum in Madrid sees
Picasso’s Guernica in front of him. As a complex unconditional stimulus, the
painting elicits unconditional visual responses—that is, neurophysiological
changes produced by stimulation (Zilio, 2010). According to Skinner (1963),
“when a man sees red, he may be seeing the physiological effect [the visual
response] of a red stimulus; when he merely imagines red, he may be seeing the
same effect [the visual response] re-aroused” (p. 957).
Now assume that the visitor of the Reina Sofía Museum also listens to
Beethoven’s Appassionata when in front of Guernica. This produces the Pavlovian
pairing of an auditory stimulus (Beethoven’s Appassionata) and a visual stimulus
(Picasso’s Guernica). As a result, every time the visitor listens to this particular
sonata, visual responses of Guernica will be elicited—an example of what Skinner
(1953/1965) called conditional visual responses. Assume further that the visitor
comes back to his home country and “tries to remember” the details of Guernica a
few months later: that is, he manipulates his environment so as to increase the
probability of responses of the class, “seeing Guernica.” He plays Beethoven’s
compositions on his stereo, read books about Picasso, and so on. He may also emit
operant visual responses (Skinner, 1953). After manipulating the environment so
as to “see” Guernica in the absence of the painting, the visitor now says: “I am
seeing Guernica.” In this case, the visual response is acting as discriminative
stimulus for the verbal description “I’m seeing Guernica” (Skinner, 1963). In this
context, Skinner is well known for distinguishing seeing something from seeing
that we are seeing it:
Someone shows you a picture of a group of scientists, among them Einstein. He
asks, "Is Einstein there?" and you say, “Yes,” as you have been taught to do in
thousands of comparable cases. But suppose he asks, “Do you see Einstein?”
and you say, “Yes.” What have you reported? Did you, in response to his
question, simply look at Einstein a second time? If so, how do you distinguish
between “seeing Einstein” and “seeing that you are seeing Einstein”? A
possibility, which needs to be considered, is that in reporting that you see
Einstein, you are reporting a response rather than a stimulus. No matter how
obscure its dimensions, the behavior called seeing must be involved, and you
must be reporting it, rather than the presence of the thing seen, when you report
that you see something. You may be reporting the same thing when you report
that you see something which “isn't really there” - when you are merely
“imagining how Einstein looked.” Seeing something in memory is not
necessarily seeing a copy. (…) When I recall how something looked, I may
simply be recalling how I once looked at something. There was no copy inside
me when I first looked at it, and there is none now. I am simply doing again
what I once did when I looked at something, and I can tell you that I am doing
so. (Skinner, 1967, pp. 329-330)
Skinner’s theory of seeing contrasts with the notion that perceptual responses
in the absence of the thing seen are cognitive and/or neural representations of the
original stimulus in the “mind’s eye” (cf. Lopes & Abib, 2002; Tourinho, 1994;
Zilio, 2010). In Skinner’s theory of perception there is no need for such copies or
representations: “what is ‘seen’ is the presentation, not a representation” (Skinner,
1985, p. 292).
Figure 3 summarizes Skinner’s model of perception. Perceptual activities are
viewed as responses (and not simply as effects caused by stimulation) because they
are sensitive to the history of organism: They can occur as a function of operant or
respondent contingencies. In this sense, they are no less behavioral than motor
responses. Also, just as motor responses of different intensities and topographies
are classified as “bar pressing” as long as they fulfill a common criterion (e.g., the
deflection of the lever), perceptual responses qualify as “seeing Guernica” (for
example) because of their shared discriminative effects (saying, “I see Guernica”).
As in the motor case, perceptual stimulation may not cause identical neural
changes each time it occurs. The responses can and do vary, but are lumped in the
same “class” when we describe what we are seeing as something “that we saw
before.” As in the motor case, there is a limit to within-class variation too.
Evolving perceptual responses can differ from the original one to such an extent
that the visitor stops saying he is seeing Guernica—he now “sees something else.”
Figure 3: Skinner’s model of perception. 1, 2, and 3 indicate an unconditional visual
relation, a conditional visual relation, and an operant visual relation, respectively.
Continuous arrows with filled and unfilled arrowheads represent relations of elicitation
and discriminative control, respectively. Dotted arrows with unfilled arrowheads represent
stimulus pairings.
Neurophysiological hypothesizing plays a complex role in Skinner’s theory of
perception. Skinner assumes that different kinds of stimuli—unconditional,
conditional, and discriminative—can produce perceptual responses of the same
class, that is, similar neurophysiological effects. He also assumes that perceptual
responses occur even when we “see something in the absence of the thing seen”
and that we can in turn respond to these responses, which are now acting as
discriminative stimuli. The plausibility and validation of Skinner’s theory of
perception ultimately depends on what neuroscience will tell us about the workings
of the brain. As it stands, recent neurophysiological findings seem to indicate that
Skinner was on the right track. Although his take on neuroscience may appear
primitive by current standards, more recent research (e.g., Borst & Kosslyn, 2008;
Cichy, Heinzle, & Haynes, 2012; Kosslyn, Thompson & Ganis, 2006; Kreiman,
Koch & Fried, 2000) has uncovered important similarities between the
neurophysiological processes related to visual perception (unconditional vision)
and “seeing in the absence of the thing seen” (also known as “mental imagery”).
Kosslyn, Thompson and Ganis (2006), for example, conclude that “visual imagery
evokes many of the same processing mechanisms used in visual perception” (p.
135). As Skinner anticipated, the perceptual responses produced by unconditional
relations and those under other kinds of stimulus control (such as conditional and
discriminative) are clearly similar to one another.
Final Thoughts
The main goal of this paper was to discuss some of Skinner’s positive (noncritical) views about physiology. A first step in the argument was to clarify the
differences, according to Skinner, between the domains of behavior analysis and
physiology as well as the role of both sciences in the explanation of behavior. I
also discussed specific examples from Skinner’s works of how physiology could
inform behavior analysis by providing explanations of motivational phenomena,
discrimination, elicitation, and the neural mechanisms of reinforcement. Close
attention was given to Skinner’s use of physiological notions in developing his
theories of private events and perception. The Skinnerian concept of privacy is
based, at least in part, on the “special contact” afforded by different physiological
systems, and his criticisms of representational theories of imagery embody a
specific behavioral/neurophysiological theory of perception.
These considerations converge toward one conclusion: Skinner never
defended an anti-physiological position on behavior. It is one thing to focus on
historical variables and to leave the study of other factors to physiology, quite
another to deny the importance of the latter. Skinner can hardly be accused of that.
In his own words: “A behavioral account is incomplete, in part because it leaves a
great deal to neurology” (1985, p. 297).
Baars, B. J., & Gage, N. M. (2010). Cognition, brain, and consciousness: Introduction to
cognitive neuroscience. London, UK: Academic Press.
Baer, D. M. (1996). On the invulnerability of behavior-analytic theory to biological
research. The Behavior Analyst, 19, 83-84.
Barnes-Holmes, D. (2003). For the radical behaviorist biological events are not biological
and public events are not public. Behavior and Philosophy, 31, 145-150.
Bartolomeo, P. (2008). The neural correlates of visual mental imagery: An ongoing debate.
Cortex, 44, 107-108.
Baum, W. M. (1994). Understanding behaviorism: Science, behavior, and culture. New
York, NY: HarperCollins.
Bickle, J. (1998). Psychoneural reduction: The new wave. Cambridge, MA: The MIT
Borst, G., & Kosslyn, S. M. (2008). Visual mental imagery and visual perception:
Structural equivalence revealed by scanning processes. Memory & Cognition, 36,
Bradnan, W. A. (1982). On behavioristic versus neurophysiologic accounts of psychotic
behavior. Journal of Medicine and Philosophy, 7, 289-303.
Branch, M. N. (2006). How research in behavioral pharmacology informs behavioral
science. Journal of the Experimental Analysis of Behavior, 85, 407-423.
Bullock, D. (1996). Toward a reconstructive understanding of behavior: A response to
Reese. The Behavior Analyst, 19, 75-78.
Campeau, S., Miserendino, M., & Davis, M. (1992). Intra-Amygdala infusion of the NMethyl-D-Aspartate receptor antagonist AP5 blocks acquisition but not expression of
fear-potentiated startle to an auditory conditioned stimulus. Behavioral Neuroscience,
106, 569-574.
Carvalho Neto, M. B. (1999). Fisiologia e behaviorismo radical: Considerações sobre a
caixa preta [Physiology and radical behaviorism: Considerations on the black box]. In
R. C. Wielenska, & R. Kerbauy (Orgs), Sobre comportamento e cognição (Vol. 4, pp.
262-271). Santo André, Brasil: ESETEC.
Carvalho Neto, M. B., & Tourinho, E. Z. (1999). Skinner e o lugar das variáveis biológicas
em uma explicação comportamental [Skinner and the place of biological variables in
behavioral explanation]. Psicologia: Teoria e Pesquisa, 15, 45-53.
Catania, A. C. (2000). From behavior to brain and back again: Book review of Orbach on
Lashley-Hebb. Psycholoquy, 11(27).
Catania, A. C. (2007). Learning (4th ed.). New York: Sloan Publishing.
Cichy, R. M., Heinzle, J., & Haynes, J.-D. (2012). Imagery and perception share cortical
representations of content and location. Cerebral Cortex, 22, 372-380.
Davis, M., & Whalen, P. J. (2001). The amygdala: Vigilance and emotion. Molecular
Psychiatry, 6, 13-34.
Donahoe, J. W. (1996). On the relation between behavior analysis and biology. The
Behavior Analyst, 19, 71-73.
Donahoe, J. W. (2002). Behavior analysis and neuroscience. Behavioural Processes, 57,
Donahoe, J. (2004). Interpretation and experimental analysis: An underappreciated
distinction. European Journal of Behavior Analysis, 5, 83-89.
Donahoe, J. W., Palmer, D. (1994). Learning and complex behavior. Boston, MA: Allyn and
Elcoro, M. (2008). Including physiological data in a science of behavior: A critical
analysis. Revista Brasileira de Terapia Comportamental e Cognitiva, 10, 253-261.
Farah, M. J. (2000). The neural bases of mental imagery. In M. S. Gazzaniga (Ed.), The
new cognitive neurosciences (pp. 965-974). Cambridge, MA: The MIT Press.
García-Hoz, V. (2004). A note on Skinner and Pavlov’s physiology. The Spanish Journal
of Psychology, 7, 153-160.
Gazzaniga, M. S., Ivry, R. B., & Mangun, G. R. (2002). Cognitive neuroscience: The
biology of the mind (2nd ed.). New York, NY: W. W. Norton & Company.
Glenn, S., & Madden, G. J. (1995). Units of interaction, evolution and replication: Organic
and behavioral parallels. The Behavior Analyst, 18, 237-251.
Greenberg, G. (1983). Psychology without the brain. The Psychological Record, 33, 49-58.
Greenberg, G., & Lambdin, C. (2007). Psychology is a behavioral science, not biological
science. A discussion of the issue and a review of neural theories of mind: Why the
mind-brain problem may never be solved, by William Uttal. The Psychological
Record, 57, 457-475.
Gulyás, B. (2009). Functional neuroimaging and the logic of conscious and unconscious
mental processes. In E. Pöppel & A. von Müller (Eds.), Neural correlates of thinking
(pp. 141-173). Berlin, GE: Springer.
Hayes, S. C., & Brownstein, A. (1986). Mentalism, behavior-behavior relations, and a
behavior-analytic view of the purposes of science. The Behavior Analyst, 9, 175-190
Horst, S. (2007). Beyond reduction: Philosophy of mind and post-reductionist philosophy
of science. New York, NY: Oxford University Press.
Illard, S. S., & Feldman, D. (2001). On mental events, disciplinary boundaries, and
reductionism: A reply to Plaud. Journal of Clinical Psychology, 57(9), 1103-1107.
Kandel, E. (1976). Cellular basis of behavior: An introduction to behavioral neurobiology.
San Francisco, CA: W. H. Freeman and Company.
Kosslyn, S. M., Thompson, W. L., & Ganis, G. (2006). The case for mental imagery. New
York, NY: Oxford University Press.
Kreiman, G., Koch, C., & Fried, I. (2000). Imagery neurons in the human brain. Nature,
408(16), 357-361.
Lacey, H. (1984). A posição de Skinner acerca da predição e controle do comportamento
[Skinner’s position on the prediction and control of behavior]. Cadernos de História e
Filosofia da Ciência, 6, 5-34.
Laurenti, C., & Lopes, C. E. (2008). Uma explicação não-causal do comportamento no
behaviorismo radical [A noncausal explanation of behavior in radical behaviorism].
Acta Comportamentalia, 16, 379-397.
LeDoux, J. (2000). Emotion circuits in the brain. Annual Review of Neuroscience, 23, 155184.
Lopes, C. E., & Abib, J. A. D. (2002). Teoria da percepção no behaviorismo radical [The
theory of perception in radical behaviorism]. Psicologia: Teoria e Pesquisa, 18, 129137.
Loucks, R. B. (1941). The contribution of physiological psychology. Psychological
Review, 48, 105-126.
Lubinski, D., & Thompson, T. (1987). An animal model of the interpersonal
communication of interoceptive (private) states. Journal of the Experimental Analysis
of Behavior, 48, 1-15.
Lubinski, D., & Thompson, T. (1993). Species and individual differences in
communication based on private states. Behavioral and Brain Sciences, 16, 627-680.
Machamer, P. (2009). Learning, neuroscience, and the return of behaviorism. In J. Bickle
(Ed.), The Oxford handbook of philosophy and neuroscience (pp. 166-176). New
York, NY: Oxford University Press.
Marr, M. J. (2003). The what, the how, and the why: The explanation of Ernst Mach.
Behavior and Philosophy, 31, 181-192.
Marr, M. J., & Zilio, D. (2013). No island entire of itself: Reductionism and behavior
analysis. European Journal of Behavior Analysis, 14, 241-257.
Mechner, F. (2008). An invitation to behavior analysts: Review of In Search of Memory:
The Emergence of a New Science of Mind by Eric R. Kandel. Journal of the
Experimental Analysis of Behavior, 90, 235-248.
Moore, J. (1980). On behaviorism and private events. The Psychological Record, 30, 459475.
Moore, J. (2001). On psychological terms that appeal to the mental. Behavior and
Philosophy, 29, 167-186.
Moore, J. (2002). Some thoughts on the relation between behavior analysis and behavioral
neuroscience. The Pyschological Record, 52, 261-279.
Moore, J. (2008). Conceptual foundations of radical behaviorism. New York, NY: Sloan
Moore, J. (2009). Why the radical behaviorist conception of private events is interesting,
relevant, and important. Behavior and Philosophy, 37, 21-37.
Moore, J. (2010). Behaviorism and the stages of scientific activity. The Behavior Analyst,
33, 47-63.
Mustaca, A. E. (2003). Análisis experimental del comportamiento y neurociencias [The
experimental analysis of behavior and the neurosciences]. Acta Colombiana de Psicología,
10, 7-22.
Natsoulas, T. (1983). Perhaps the most difficult problem faced by behaviorism.
Behaviorism, 11(1), 1-26.
Natsoulas, T. (1985). The treatment of conscious content: Disorder at the heart of radical
behaviorism. Methodology and Science, 18, 81-103.
Paniagua, F. A. (1985). The relational definition of reinforcement: Comments on
circularity. The Psychological Record, 35, 193-202.
Panksepp, J. (1990). Can “mind” and behavior be understood without understanding the
brain? A response to Bunge. New Ideas in Psychology, 8, 139-149.
Palmer, D. C. (2009). The role of private events in the interpretation of complex behavior.
Behavior and Philosophy, 37, 3-19.
Prado Jr., B. (1982). Breve nota sobre o operante: Circularidade e temporalidade [A short
note on the operant: Circularity and temporality]. In B. Prado Jr. (Org.). Filosofia e
comportamento (pp. 110-118). São Paulo, Brasil: Brasiliense.
Razran, G. (1965). Russian physiologists’ psychology and American experimental
psychology: A historical and a systematic collation and a look into the future.
Psychological Bulletin, 63, 42-64.
Reese, H. W. (1996). How is physiology relevant to behavior analysis? The Behavior
Analyst, 19, 61-70.
Sah, P., Westbrook, R. F., & Lüthi, A. (2008). Fear conditioning and long-term
potentiation in the amygdala: what really is the connection? Annals of the New York
Academy of Sciences, 1129(1), 88-95.
Schaal, D. W. (2003). Explanatory reductionism in behavior analysis. In K. A. Lattal, & P. N.
Chase (Eds.), Behavior theory and philosophy (pp. 83-102). New York, NY: Kluwer
Schaal, D. W. (2005). Naming our concerns about neuroscience: A review of Bennett and
Hacker’s Philosophical Foundations of Neuroscience. Journal of the Experimental
Analysis of Behavior, 84, 683-692.
Schaffner, K. F. (1967). Approaches to reduction. Philosophy of Science, 34, 137-147.
Schick, K. (1971). Operants. Journal of Experimental Analysis of Behavior, 15, 413-423.
Schnaitter, R. (1984). Skinner on the "mental" and the "physical". Behaviorism, 12(1), 114.
Shepherd, G. M. (1994). Neurobiology (3rd ed.). New York, NY: Oxford University Press.
Silva, M. T. A. (1987). Behaviorismo e neurociências [Behaviorism and the
neurosciences]. In D. G. de Souza, V. R. L. Otero, & Z. Alves (Eds.), Anais da XVII
Reunião Anual de Psicologia da SPRP (pp. 549-553). Ribeirão Preto, SP, Brasil.
Silva, M. T. A., Gonçalves, F. L., & Garcia-Mijares, M. (2007). Neural events in the
reinforcement contingency. The Behavior Analyst, 30, 17-30.
Skinner, B. F. (1950). Are theories of learning necessary? The Psychological Review,
57(4), 193-216.
Skinner, B. F. (1957). Verbal Behavior. New York, NY: Appleton-Century-Crofts.
Skinner, B. F. (1961a). Current trends in experimental psychology. In B. F. Skinner,
Cumulative Record: A selection of papers (2nd ed., pp. 223-241). New York, NY:
Appleton-Century-Crofts. (Original work published 1947).
Skinner, B. F. (1961b). Psychology in the understanding of mental disease. In B. F.
Skinner, Cumulative record: A selection of papers (2nd ed., pp. 194-202). New York,
NY: Appleton-Century-Crofts. (Original work published 1957).
Skinner, B. F. (1961c). The concept of reflex in the description of behavior. In B. F.
Skinner, Cumulative record: A selection of papers (2nd ed., pp. 319-346). New York,
NY: Appleton-Century-Crofts. (Original work published 1931).
Skinner, B. F. (1961d). The experimental analysis of behavior. In B. F. Skinner,
Cumulative Record: A selection of papers (2ª ed., pp. 100-131). New York, NY:
Appleton-Century-Crofts. (Original work published 1957).
Skinner, B. F. (1961e). What is psychotic behavior? In B. F. Skinner, Cumulative record: A
selection of papers (2nd ed., pp. 202-219). New York, NY: Appleton-Century-Crofts.
(Original work published 1956).
Skinner, B. F. (1963). Behaviorism at fifty. Science, 140(3570), 951-958.
Skinner, B. F. (1965). Science and human behavior. New York, NY: The Free Press.
(Original work published 1953).
Skinner, B. F. (1966a). The behavior of organisms: An experimental analysis. New York,
NY: Appleton-Century-Crofts. (Original work published 1938).
Skinner, B. F. (1966b). What is the experimental analysis of behavior? Journal of the
Experimental Analysis of Behavior, 9, 213-218.
Skinner, B. F. (1967). The problem of consciousness–A debate. Philosophy and
Phenomenological Research, 27, 325-337.
Skinner, B. F. (1969a). Contingencies of reinforcement: A theoretical analysis. New York,
NY: Appleton-Century-Crofts.
Skinner, B. F. (1969b). The machine that is man. Psychology Today, 2, 20-25, 60-63.
Skinner, B. F (1971a). A behavioral analysis of value judgments. In E. Tobach, L. R.
Aronson, & E. Shaw (Eds.), The biopsychology of development (pp. 543-551). New
York, NY: Academic Press.
Skinner, B. F. (1971b). Beyond freedom and dignity. New York, NY: Alfred A. Knopf.
Skinner, B. F (1972). Humanism and behaviorism. The Humanist, 32, 18-20.
Skinner, B. F (1973). Answers for my critics. In H. Wheeler (Ed.), Beyond the punitive
society (pp. 256-266). San Francisco, CA: Freeman.
Skinner, B. F. (1974). About behaviorism. New York, NY: Alfred A. Knopf.
Skinner, B. F. (1975a). The shaping of phylogenic behavior. Journal of the Experimental
Analysis of Behavior, 24, 117-120.
Skinner, B. F. (1975b). The steep and thorny way to a science of behavior. American
Psychologist, 30, 42-49.
Skinner, B. F. (1983). Can the experimental analysis of behavior rescue psychology? The
Behavior Analyst, 6, 9-17.
Skinner, B. F. (1985). Cognitive science and behaviourism. British Journal of Psychology,
76, 291-301.
Skinner, B. F. (1986). Is it behaviorism? The Behavioral and Brain Sciences, 9, 716.
Skinner, B. F. (1987). Whatever happened to psychology as the science of behavior?
American Psychologist, 42, 780-786.
Skinner, B. F. (1988). Comments. In A. C. Catania & S. Harnad (Eds.), The selection of
behavior: The operant behaviorism of B. F. Skinner: Comments and consequences.
New York, NY: Cambridge University Press.
Skinner, B. F. (1989a). Genes and behavior. In B. F. Skinner, Recent issues in the analysis
of behavior (pp. 49-56). Columbus, OH: Merrill Publishing Company.
Skinner, B. F. (1989b). The behavior of organisms at 50. In B. F. Skinner, Recent issues in
the analysis of behavior (pp. 121-136). Columbus, OH: Merrill Publishing Company.
Skinner, B. F. (1989c). The operant side of behavior therapy. In B. F. Skinner, Recent
issues in the analysis of behavior (pp. 73-84). Columbus, OH: Merrill Publishing
Company. (Original work published 1988).
Skinner, B. F (1989d). The origins of cognitive thought. American Psychologist, 44, 13-18.
Skinner, B. F. (1990). Can psychology be a science of mind? American Psychologist, 45,
Skinner, B. F (1993). A world of our own. Behaviorology, 1, 3-5.
Skinner, B. F. (2009). Conversation between B. F. Skinner and E. O. Wilson. In P. Naour,
E. O. Wilson and B. F. Skinner: A dialogue between sociobiology and radical
behaviorism (pp. 61-85). New York, NY: Springer.
Smith, L. D. (1992). On prediction and control: B. F. Skinner and the technological ideal of
science. American Psychologist, 47(2), 216-223.
Staddon, J. (2001). Adaptive Dynamics: The theoretical analysis of behavior. Cambridge,
MA: The MIT Press.
Staddon, J., & Bueno, J. L. O. (1991). On models, behaviorism and the neural basis of learning.
Psychological Science, 2, 3-11.
Starling, R. R. (2000). A interface comportamento/neurofisiologia numa perspectiva
behaviorista radical: O relógio “causa” as horas? [The behavior/neurophysiology
interface from a radical behaviorist perspective: Does a clock “cause” the hours?]
In R. R. Kerbauy (Org.), Sobre Comportamento e Cognição (vol. 5, pp. 3-15).
Santo André, Brasil: ESETEC Ed. Associados.
Strumwasser, F. (1994). The relations between neuroscience and human behavioral
science. Journal of the Experimental Analysis of Behavior, 61(2), 307-317
Thompson, T. (2007). Relations among functional systems in behavior analysis. Journal of
the Experimental Analysis of Behavior, 87(3), 423-440.
Thompson, T. (2008). Funcional analysis of behavior and neuroscience. In J. E. Burgos &
E. Ribes-Iñesta (Eds.), The brain-behavior nexus: Conceptual issues: Proceedings of
the 10th biannual symposium on the science of behavior (pp. 85-100). Guadalajara,
México: Universidad de Guadalajara.
Timberlake, W., Schaal, D. W., & Steinmetz, J. E. (2005). Relating behavior and
neuroscience: introduction and synopsis. Journal of the Experimental Analysis of
Behavior, 84, 305-311.
Tonneau, F. (2008). The concept of reinforcement: Explanatory or descriptive? Behavior and
Philosophy, 36, 87-96.
Tourinho, E. Z. (1994). A noção pragmatista de conhecimento e a noção skinneriana de
conhecimento de si mesmo [The pragmatist notion of knowledge and the Skinnerian notion
of self-knowledge]. Acta Comportamentalia, 2, 219-232.
Tourinho, E. Z. (2006a). On the distinction between private events and the physiology of
the organism. The Behavior Analyst Today, 7(4), 548-559.
Tourinho, E. Z. (2006b). Private stimuli, covert responses, and private events: Conceptual
remarks. The Behavior Analyst, 29, 13-31.
Tourinho, E. Z., Teixeira, E., & Maciel, J. M. (2000). Fronteiras entre análise do
comportamento e fisiologia: Skinner e a temática dos eventos privados [Boundaries
between behavior analysis and physiology: Skinnner and the topic of private events].
Psicologia: Reflexão e Crítica, 13, 425-434.
Woodward, J. (2003). Making things happen: A theory of causal explanation. New York,
NY: Oxford University Press.
Zilio, D. (2010). A natureza comportamental da mente: Behaviorismo radical e filosofia da
mente [The behavioral nature of the mind: Radical behaviorism and philosophy of
mind]. São Paulo: Editora Cultura Acadêmica.
Zilio, D. (2013). Análise do comportamento e neurociências: Em busca de uma possíve
síntese [Behavior analysis and the neurosciences: In search of a possible synthesis].
Tese de Doutorado, Instituto de Psicologia, Universidade de São Paulo, São Paulo.
Zilio, D., & Dittrich, A. (2014). O que fazer com os eventos privados? Reflexões a partir
das ideias de Baum, parte I: A definição de privacidade [What to do with private
events? Reflecting on Baum’s ideas, part I: The definition of privacy]. Acta
Comportamentalia, 22, 483-496.
Zuriff, G. E. (1985). Behaviorism: A conceptual reconstruction. New York, NY: Columbia
University Press.
Author’s Note
Writing this article was supported by FAPESP - São Paulo Research Foundation
(2009/18324-1/2013/17950-1). The article itself is based on a talk presented at the
1st Amazonian Simposium on Behavior Research and Theory (Universidade
Federal do Pará, Belém, Brasil, November 2013). I would like to thank Maria
Helena Leite Hunziker and François Tonneau for their helpful comments on earlier
drafts. Correspondence concerning this article should be addressed to Diego Zilio,
Federal University of Espírito Santo (UFES)- Department of Social and
Developmental Psychology, Fernado Ferrari Ave, 514. Goiabeiras. Vitoria- ES.
Brazil. 29075-910