Download Settling The Stimulus-Substitution Issue Is A Prerequisite For Sound

Back to Psychophysiological
Home Page
Behavioral And Brain Sciences (1993), 16:2, pp.392-398
Malmo_Furedy93.doc
Continuing Commentary
Commentary on Jaylan Sheila Turkkan (1989) Classical conditioning: The new hegemony. BBS 12:121-179.
Abstract of the original article: Converging data from different disciplines are showing the role of classical conditioning processes in the
elaboration of human and animal behavior to be larger than previously supposed. Restricted views of classically conditioned responses as merely
secretory, reflexive, or emotional are giving way to a broader conception that includes problem-solving and other rule-governed behavior thought to
be the exclusive province of either operant conditioning or cognitive psychology. These new views have been accompanied by changes in the way
conditioning is conducted and evaluated. Data from a number of seemingly unrelated phenomena such as relapse to drug abuse by post addicts, the
placebo effect, and the immune response appear to involve classical conditioning processes. Classical conditioning, moreover, has been found to
occur in simpler and simpler organisms and recently even demonstrated in brain slices and in utero. Several research areas that have used the
classical conditioning process as an explanatory model are integrated; teleological interpretations of the classically conditioned CR are challenged
and some basic principles for testing conditioning in diverse areas are offered.
Settling the stimulus-substitution issue is a
prerequisite for sound nonteleological neural analysis
of heart-rate deceleration conditioning
Robert B. Malmo a and John J. Furedy b "Neuropsychology Laboratory,
Department of Psychiatry, McGill University, Montreal, Quebec, Canada H3A 1A1
and "Department of Psychology, University of Toronto, Toronto, Ontario, Canada
M5S 1AL
In discussing the place of Pavlovian conditioning in psychology,
Rescorla (1988) called attention to the currently intense interest of
neuroscientists in the neural basis of conditioning, whose analysis
Rescorla believes should be guided by the kind of teleological
assumptions that Furedy (1989) argued against in commenting on
Turkkan’s (1989) target article. In particular, Rescorla characterized
Pavlovian conditioning as "the learning of relations among events so as
to allow the organism to represent its environment" (Rescorla 1988, p.
151, our emphasis). The chief purpose of our commentary is to call
attention to neuropsychological conditioning experiments by Malmo in
which the UR (centrally elicited heart-rate deceleration in rats) appears to
fit into the classical nonteleological and noncognitive S-R model that
Rescorla (1988) rejects.
Malmo (1963; 1965) developed a rat preparation that appears optimal
for classical conditioning of heart-rate (HR) deceleration. Malmo's
method incorporated the following features:
(1) His US (lateral septal stimulation) is favorable in a number of
respects: (a) the elicited HR deceleration is large-magnitude and
unidirectional, avoiding initial HR acceleration, which is encountered
with stimulation of the medial septal nucleus (Malmo 1964; Ross &
Malmo 1979); (b) with appropriate electrode placement in the lateral
septal area the short-latency HR deceleration is elicited directly instead
of being secondary to a change in mean arterial pressure (see Ross &
Malmo 1979) (in this respect the rabbit appears to differ from the rat in
not exhibiting this desirable feature; see Van Dercar et al. 1970); (c) the
UR is nonhabituating (unlike shock and loud-noise USs); and (d) the
cardiac deceleration is not secondary to a change in respiration.
(2) The CS is an exteroceptive (4K Hz tone) stimulus; when presented
alone it has no effect on HR.
(3) The relatively short 2.5 sec CS-US interval stands in contrast to the
8-10 sec US intervals used in some more recent studies by cognitively
oriented workers. However, the 2.5 sec CS-US interval permits the
measurement of the (anticipatory) CR on all reinforced trials, thereby
avoiding complete dependence on CS-alone test trials to assess
conditioning (Malmo 1963, pp 28-29).
(4) The results (Malmo 1963, Fig. 11; see also Malmo 1965) yield
highly orderly acquisition and extinction functions. As in the human
negative-tilt preparation, the CR and UR are in the same (decelerative)
direction (Furedy et al. 1989).
(5) Various observations rule out nonassociative factors such as
392
BEHAVIORAL AND BRAIN SCIENCES (1993) 16:2
sensitization or pseudoconditioning. For example, there is no HR
response to tone even after at least six sessions of brain stimulation
alone.
The scant attention paid to these remarkable findings is probably
accounted for by the cognitive shift that was gathering force at the time
they were published. Six years later, however, Malmo's (1965)
conditioning study was referred to as the foundation for HR decelerative
conditioning experiments with rabbits in Schneiderman's laboratory
(Sideroff et al. 1971). For these investigators (and for others later on)
there appears to be a compulsion (perhaps encouraged by the operant
reinforcement properties of some of the US brain areas used) to find
some feature, other than mere UR elicitation, that is indispensable for
conditioning. We, on the other hand, believe it possible that for
conditioning the only essential feature of brain stimulation may be its
elicitation of HR deceleration. To the best of our knowledge, this
possibility has not been considered recently. Investigators of HR
classical conditioning in animal neuropsychological preparations appear
to assume that the US must be "motivationally potent.'
Having refuted the conclusion of Sideroff et al. (1971) that the US
must be aversive to be effective, Sideroff et al. (1972) still expressed the
view "that a motivationally potent US is necessary for establishing HR
conditioning." They gauged motivational potency of the US in terms of
its efficacy as an operant rein-forcer; and they attributed the failure of
Sideroff et al. (1971) to obtain conditioning using low-intensity septal
USs (which were nonaversive) to the presumed lower efficacy of septal,
compared with hypothalamic, stimulation for intracranial selfstimulation (ICSS). Sideroff et al. (1972) used hypothalamic stimulation.
This interpretation appears unparsimonious; and worse still, as we shall
argue, it ignores a central issue concerning classical conditioning. No
convincing evidence was presented to indicate that the low-intensity
septal USs were as effective in eliciting short-latency, nonhabituating,
large-amplitude cardiac decelerations as the high-intensity ones.
Malmo's (1965) findings, which were not referred to by Sideroff et al.
(1972), strongly suggest that it is the efficacy of UR elicitation rather
than some vague "motivational" factor that is the crucial variable.
Malmo (1965) found that the correlation between mean magnitude CRs
(HR decelerations) and mean number of bar presses in ICSS was only
.03. Moreover, subjects 5 and 8, who from the midway point on to the
end of acquisition clearly stood out (with two others) as showing the
largest CRs, could not be trained to self-stimulate. They showed no
aversive reactions to the septal stimulation, which seemed to be motivationally neutral. From the findings of Gardner and Malmo (1969, Figure
1, p. 67) it appears that Malmo's (1965) stimulation sites in the septal
area were dorsal to aversive septal sites that, contrary to the supposition
of Sideroff et al. (1971), do not require higher current for escape than the
dorsal stimulation sites require for approach. In short, the USs for those
two superb conditioners appeared to be devoid of any operant
reinforcement properties.
These important leads from Malmo's laboratory have not been
followed up. Instead, in neuropsychological studies, there has been a
shift away from brain-stimulation USs to cutaneous shock USs with
aversive properties (Hitchcock et al. 1989; Kapp et al. 1979). Left
unresolved is the crucial issue of stimulus substitution by contiguity (in
the absence of other reinforcement properties). We believe that the
resolution of this central issue is a prerequisite for understanding the
neural mechanisms underlying classical conditioning of HR deceleration.
Our view is therefore contrary to the currently popular ideological
approach exemplified in Rescorla's (1988) paper, which suggests that the
nonteological, stimulus-substitution, S-R approach to Pavlovian
conditioning is outmoded on scientific, evidentiary grounds. The
findings from Malmo's laboratory have admittedly not been widely cited
in the more "modern" treatments of conditioning, but as has been shown
above, they are logically relevant for problems in the conditioning of
brain-stimulation-elicited HR deceleration in animal preparations. There
is also a parallel with similar issues in human autonomic conditioning
where we believe S-R factors should be carefully considered along with
S-S factors in a dual-process approach (see Furedy & Riley 1987), in
particular, where one aim is the applied one of inducing the learning of a
potentially medically useful target CR such as HR deceleration: S-R,
response-learning factors need to be "remembered" (Furedy 1979), a
position that contrasts with focusing solely on conditioning as "the
learning of relations among events" (Rescorla 1988, p. 151).
Classical conditioning: The hegemony is not
ubiquitous
Harald Merckelbach and Marcel van den Hout Department of Mental
Health Sciences, University of Umburg, 6200 MD Maastricht, The Netherlands
Electronic mall: [email protected]
In the past years, classical conditioning has gotten bad press, as noted,
for example, by Davey (1987). Turkkan's (1989) excellent review does
much to correct this bad reputation. It shows that the Pavlovian tradition
provides fruitful procedures for studying such (seemingly) unrelated
phenomena as taste aversion, immunosuppression, and drug dependency.
One can use, as Turkkan chooses to do, a quasipolitical term, in casu
"hegemony," to describe this happy state of affairs.
Not all areas of interest to behavioral researchers are characterized by
a hegemony of classical conditioning, however. For example, in the field
of anxiety disorders, "traditional" classical conditioning accounts are on
their way out (e.g., Marks 1977; Rachman 1977). This is all the more
remarkable when one realizes that the first generation of behavior
therapists were (and still are; Wolpe 1989) convinced that most clinical
phobias and panic disorders are conditioned fear responses from
confronting an unconditioned, painful event (e.g., Wolpe & Rachman
1960). Yet subsequent research has failed to find clear CS-UCS occurrences in the learning history of most anxiety disorder patients (e.g.,
McNally & Steketee 1985; Merckelbach et al. 1989a).
In an attempt to overcome this and other problems, (e.g., the
selectivity of fears; Costello 1982), a number of authors have tried to
revise the Pavlovian theory of pathological fear. There is, for example,
Seligman's preparedness hypothesis (1971), in which Pavlovian and
Darwinian concepts were combined, and Eysenck's incubation theory
[see Eysenck: "The Conditioning Model of Neurosis" BBS 2(1) 1979], in
which the Napalkov effect (i.e., conditioned fear can act as a reinforcing
event) is stressed. Reviews of the experimental (McNally 1987) and
clinical (e.g., Merckelbach et al. 1989b) studies, however, have
concluded that there is only meager support for an evolutionary
preparedness factor operating in the etiology of human fears. Similarly,
classical conditioning experiments that have sought to confirm the
Napalkov phenomenon have yielded disappointing results (e.g.,
Nicholaichuk et al. 1982).
It is no wonder, then, that researchers lose their faith in concepts such
as preparedness and incubation, while cognitive theories (Eysenck 1988)
are gaining influence. This is not to say that an empirically satisfying
conditioning theory is impossible. One striking feature of preparedness
as well as incubation theory is that neither made use of modern
conditioning notions such as UCS inflation, latent inhibition, and
blocking. It may well be that powerful theories emerge once these
notions are incorporated in conditioning interpretations of anxiety (e.g.,
Davey 1989; Reiss 1980). These are future promises, however. For the
moment, there is no such thing as a hegemony of classical conditioning
in the field of anxiety disorders.
Stimulus correlations in complex operant settings
Francois Tonneau
Department of Psychology, Temple University, Philadelphia, PA 19122
In her interesting target article, Turkkan (1989) effectively conveys the
impression of a "classical conditioning hegemony" by taking up a datadriven approach and by reviewing a wide range of experimental results.
The many advantages of such empirical approaches, however, should
not make us forget the existence of various conceptual constraints on
data selection and presentation. With regard to the target article, one's
basic conception of behavioral explanation (e.g., Bolles 1984) necessarily influences the kind of "classical conditioning hegemony" to be
proposed, the way in which the hegemony unifies experimental data, and
the direction in which the hegemony could be further extended.
Consider Turkkan's discussion of the relationships between classical
conditioning (hereafter CC) and "memory," for example. Recall behavior
is assimilated to the production of conditional responses (p. 133); it is
then shown that recall can be related to temporal/ordering factors and
variables such as intensity or contingency in ways that parallel CC
phenomena. Does this exemplify some CC "determinants' of memory, as
Turkkan puts it (p. 133)? Perhaps, but the data could just as well be
taken to imply that memory processes determine CC. Indeed, in
memory-based accounts of Pavlovian responding it is assumed that
conditional responses result from the retrieval of an unconditional
stimulus representation (Bolles 1985; Pearce & Hall 1980; Rescorla
1974; Wagner 1981). How the formal similarities reviewed by Turkkan
(pp. 133-35) are to be interpreted thus clearly depends on one's
underlying theoretical and conceptual choices: From a cognitive
standpoint (e.g., Dickinson 1980), CC just represents a set of surface
regularities and cannot serve as a basic behavioral explanation; CC is
itself in need of explanation and assumes a derived status. Turkkan's
reliance on CC as an explanatory tool is perhaps more readily understood
in the context of a "functional analysis" (Wixted & McDowell 1989; see
also Wixted 1989) that does not focus on mediating processes but aims
at establishing wide-ranging relationships between independent and
dependent variables.
More generally, the target article (as well as the accompanying
commentaries) can be examined in the light of Hineline's (1986; 1990)
recent discussions of "organism-based" and "environment-based" modes
of psychological theorizing. Especially relevant is Hineline's remark that
the two explanatory modes entail different conceptions of psychological
process: "For the organism-based position, process is what goes on
inside the organism; process is said to mediate, or underlie behavior.
From the behavior-analytic viewpoint, process is the very interplay
between environment and behavior" (1986, p. 62). My
BEHAVIORAL AND BRAIN SCIENCES (1993) 16:2 393
point is that both conceptions of psychological process can be found in
Turkkan's paper, whereas many of the commentaries seem to have been
written from an organism-based viewpoint. For example, an important
theme of the target article appears to be the search for laws relating
behavior to prior stimulus correlations (cf. the "litmus tests" 1-6, p. 123):
This theme defines the possible field of a "CC hegemony" and accords
well with environment-based psychological theory. On the other hand,
CC results are attributed not to the stimulus correlations themselves
(Skinner 1977) but to internal associative processes (pp. 134-35) - a
cognitive explanation (see Schnaitter 1986, pp. 260-61) which could
make sense only in the context of the "organism-based" theories
(Dickinson 1980; Dickinson & Mackintosh 1978) that Turkkan criticizes
for their heavy use of "hypothetical intervening variables" (p. 124). Thus
it is not clear that Turkkan's simultaneous use of different explanatory
modes leads to a consistent cognitive/behavioral integration, the
feasibility of which remains debated (see Catania 1984; Hineline 1984;
Lee 1990; Shimp 1989; Williams 1984). On the contrary, Turkkan's
strategy has surely resulted in some confusions concerning the target
article and may even explain the criticisms evident in some of the
subsequent commentaries (e.g., C. Fields 1989; Jacobs 1989; on this
point see Hayes et al. 1988, p. 106-7).
Having detailed what is in my view the main conceptual problem with
Turkkan's work, I can now take up the environment-based standpoint
and try to extend one of the many interesting aspects of the target article
- the analysis of CC processes as they relate to complex operant
behavior (pp. 133-35). The remainder of my commentary is specifically
devoted to this topic.
As noted by Turkkan, in CC the operation that "strengthens"
conditional responses is a contingent relationship, or correlation,
between stimuli.l Now, whereas the traditional view of CC (e.g., Keller
& Schoenfeld 1950, pp. 15-35) emphasizes the effects of stimulus
correlations on respondent behavior exclusively, Turkkan (p. 123)
leaves open the possibility that CC effects are much more pervasive:
Environmental correlation may modify various behavioral functions of
stimuli (cf. Staats 1968) and thus affect operant as well as respondent
behavior.
In support of this thesis, Table 1 presents stimulus functions in a more
or less conventional way, from "elicitation" to various forms of
contextual control, along with possible evidence of "strengthening" by
CC relationships. It is clear that stimulus correlations can promote the
emergence (or "transfer") of elicitation, Pavlovian occasion-setting,
Pavlovian reinforcement, and operant reinforcement (see also Brown
1985; Dinsmoor 1985; Nevin 1983).
Furthermore, to the extent that CC correlations "transfer" a whole
range of behavioral functions from one stimulus to another, they
constitute an important source of functional equivalence (cf.
Goldiamond 1966; see Rachlin et al. 1986, p. 38, for a stimulating
application of CC-based functional equivalence to the theory of choice
behavior). 2 In this respect, it should be noted that most of the recent
studies of functional equivalence in humans (e.g., Lazar 1977; Lazar &
Kotlarchyk 1986) use a matching-to-sample methodology that involves
various stimulus
Table 1 (Tonneau). Classical conditioning-related emergence of stimulus
functions
Stimulus function
Representative study
elicitation
Pavlovian reinforcement
operant reinforcement
occasion-setting
discriminative control
contextual control
Jenkins & Moore (1973)
Rashotte (1981)
Fantino (1977)
Holland & Forbes (1982)
Boelens & Smeets (1990) ?
Gatch & Osborne (1989) ?
394
BEHAVIORAL AND BRAIN SCIENCES (1993) 16:2
pairings (cf. Hayes et al. 1987, p. 364). Once the subjects have been
taught to match stimuli in a consistent way, the pairings between
samples and comparisons may produce a CC-based transfer of stimulus
functions: Indeed, it has been observed that matching-to-sample
promotes the transfer of discriminative and contextual control (Table 1;
see also Catania et al. 1988; de Rose et al. 1988). Thus, the transfer of
these functions in matching-to-sample studies may parallel the transfer
of stimulus functions like occasion-setting in "pure" CC experiments
(Table 1). The experimental analysis of such questions has just begun,
however, and despite promising results (Boelens & Smeets 1990), the
exact role of CC in the emergence of matching-to-sample functional
equivalence effects is not yet known (see Hall 1988) 3 -hence my
question marks following some studies in Table 1. On the whole, the
results suggest a widespread influence of CC on the
transfer/strengthening of stimulus functions (Table 1), but more
experimentation along the lines of Boelens and Smeets's (1900) study is
needed to confirm this "hegemony" of environmental correlations.
To sum up: However CC itself is conceived (Tonneau 1990), the
hypothesis that stimulus correlations underlie a wide range of functional
equivalence effects deserves careful scrutiny (cf. Boelens & Smeets
1990). These correlations may work in concert with other behavioral
processes to produce the observed transfer of function. In any case,
Turkkan's target article is an excellent occasion for intensifying CCrelated research in complex operant settings; given the recent
development of behavior-analytic alternatives to psychological
associationism (L. Fields et al. 1984; 1990), the ensuing benefits could
be numerous.
NOTES
1. Although the precise meaning of "correlation" remains to be
worked out (Williams 1983), effective CC relationships clearly include
much more than simple contiguity (cf., for example, the relative waiting
time hypothesis of Jenkins et al. 1981). The concept of correlation is
accordingly used in this commentary to emphasize the whole temporal
range of stimulus relationships that affect behavior. Note that contiguity
can always be saved by appealing to "representations" (e.g., Holland
1981; Kehoe et al. 1987; see Branch 1977; Marr 1983; Tonneau 1990).
2. Functional equivalence (the idea that two stimuli are functionally
equivalent to the extent that they control similar responses) is not to be
confused with Sidman and Tailby’s (1982) notion of stimulus
equivalence (see Hayes 1989). The present commentary addresses only
functional equivalence.
3. Also see Dube et al. (1987) for a discussion of the possible
relationships between CC and the reinforcer-specific effects observed in
matching-to-sample studies.
Author's Response
Functions and effects of Pavlovian stimuli
Jaylan Sheila Turkkan
Division of Behavioral Biology, Department of Psychiatry and Behavioral
Sciences, The Johns Hopkins University School of Medicine, Baltimore,
MD 21224
Electronic mail: [email protected]
The original target article (Turkkan 1989) questioned the widely
held view that the classically conditioned response must provide
a physiological or behavioral service to the organism for
successful conditioning to occur. Malmo & Furedy dig a bit
deeper and wider than this general nonteleological stance by
further suggesting that the unconditional stimulus (US) need not
function as a pun-
isher or a reinforcer in other contexts for it to be effective in
classical conditioning. That is, the US can be "motivationless," 1
a notion that was explored in some early review articles (e.g.,
Terrace 1973), but that has rarely been put to scientific test
because of the difficulty in identifying strong stimuli that are
ineffective in operant paradigms. From a historical perspective,
identical
electric
shock
intensities
and
food
deprivation/magnitude levels have been effectively used across
both operant and classical conditioning paradigms (see the many
studies described in Hull [1943] without regard to Skinnerian or
Pavlovian orientation). It makes intuitive sense to use a stimulus
that has some strength for a strengthening operation, and it is
known that intense stimuli often produce approach,
escape/avoidance behavior, and physiological changes. In truth,
one can question whether any suprathreshold stimulus can be
considered truly neutral.
Malmo & Furedy state that interoceptive lateral septal
stimulation is neutral and ineffective for maintaining operant selfstimulation but that it is' effective as a US for bradycardic
classical conditioning (Malmo 1965). They suggest further that
this model was not pursued because of cognitive biases against
stimulus-response (S-R) learning in favor of stimulus-stimulus
(S-S) learning. But the cognitive Zeitgeist may have no bearing
on why Malmo's heart-rate deceleration preparation was not
pursued. Consider first the issue of practicality - the technology
of administering exteroceptive shock, airpuffs to the eye, or
administering food is simply less complex and less prone to error
than stimulating the brain, which requires electrode locations and
stimulation intensities to fall within narrow and critical limits (cf.
Malmo 1964). There is also face validity in the use of the more
standard USs as reflective of the realities of appetite and pain.
But most important, Malmo's (1965) study - in part a test of
whether the septal area that he was stimulating was operantly
effective - was not conclusive because of the procedures that
were used. Rats were initially trained with a classical
conditioning procedure in which a tone was followed by septal
stimulation as a US. Note, however, that the same rats were
subsequently tested in an operant paradigm in which the septal
stimulation was administered contingent on level-presses. Astute
readers will immediately note that this is the very procedure that
has been shown to produce "learned helplessness" (e.g., Seligman
& Maier 1967) in a variety of species including rats. That is,
initial experience with a US during classical conditioning retards
later operant learning with use of the same stimulus. It is
therefore not surprising that the two rats that showed the largest
CRs during classical conditioning were the very rats who later
failed to self-stimulate operantly! They had learned only too well,
it seems, that their behavior bears no relation to the probability of
occurrence of impending events, whether positive or aversive.
Other rats in the same study did learn to self-stimulate, however,
perhaps because their stimulation intensities were in some cases
varied between the classical conditioning and the later operant
self-stimulation phase of the study. Oddly, Malmo & Furedy’s
current commentary directly contradicts the conclusions Malmo
reached in the 1965 paper wherein he stated: "The evidence
clearly indicates that the same stimulating voltage can serve as an
effective US in classical HR conditioning and also [emphasis
added] as an effective reinforcer
for intracranial self-stimulation" (p. 5). Here he states that septal
stimulation did not work as a reinforcer for intracranial selfstimulation in those studies. For some animals it did, and for
some animals it did not. Studies of this nature are best done with
separate groups of animals who experience either classical or
operant paradigms so that potential crossover effects such as
learned helplessness are avoided. Stimulating the lateral septal
area may indeed be an intriguing "motivationless" US, but the
experiments that Malmo & Furedy bring forward do not support
their claim. Finally, the medical usefulness that Malmo & Furedy
ascribe to transient bradycardic heart-rate reactivity is not
supported by clinical practice, which is exclusively directed
toward reduction of blood pressure for cardiovascular health.
I can only second Malmo & Furedy's sentiment that the
cognitive view appears to have led to a disappearance of the
response (cf. also Gormezano et al. 1983). As a result,
psychology has diverged from the only dependent measure that
can be directly measured - the overt behavior of the organism.
We now read about organisms down to the level of invertebrates
representing, processing, planning, imagining, evaluating,
remembering, correlating, and, yes, associating, rather than
behaving. What the organism does after all these ruminations has
been left out of the account. This is not a resurgence of dualism,
as has been claimed (e.g., Denny 1986), but an unequivocal
monism of the mind.
Merckelbach & van den Hout's commentary regarding the
etiology of phobias is puzzling because they also appear to
contradict their own conclusions from an earlier study
(Merckelbach et al. 1989a). Ninety-one phobic outpatients were
administered a questionnaire that classified their phobia as having
resulted either from a conditioning experience, vicarious
learning, informational learning, or a mixture of these
"pathways." Their results showed that conditioning experiences
are common among phobic patients, with the overall frequency of
conditioning experiences at 78%. Accordingly, in that discussion
Merckelbach et al. (1989a) state "the present findings . . . seem to
cast doubt on the view that conditioning events play a negligible
role [emphasis added] in the etiology of phobias" (p. 660).
Merckelbach & van den Hout's current contradictory assertion
that conditioning plays a minor role in the etiology of phobia may
be explained, however, by their definition of a "conditioning
event" in that earlier paper. For those phobias categorized as
"conditioned" they rarely found pure instances in which an
explicit event (e.g., actual harm or pain) led to a phobia about
stimuli or environments associated with that event. Instead, they
found that patients reported experiencing a first panic attack
consisting of frightening bodily sensations in a specific context
and were subsequently phobic in that context. For a conditioning
explanation to make any sense, a conditioning trial must have
occurred. As they rightly ask in that study, "Why do first panics
occur . . .?" (emphasis added).
The last three decades have seen an explosion in the number of
studies examining the relationship between adverse life
experiences and the prevalence of disease, including psychiatric
disorders (e.g., Dohrenwend et al. 1982). Life-event studies have
almost invariably used retrospective designs in which patients'
self-reported life experiences are correlated with the latency of
onset,
BEHAVIORAL AND BRAIN SCIENCES (1993) 16:2 395
severity, and prevalence of the disorder in question, as was done
in Merckelbach et al. (1989a). Such study designs have inherent
difficulties, however, that allow no easy conclusions about the
relationship between reported adverse experiences and the
development of phobia. For example, were the reported "first"
panic attacks in fact the first, since panic experiences in
childhood may either be forgotten or dismissed? Moreover, in
Merckelbach et al. (1989a), patients may have been more likely
to ascribe their seemingly irrational phobic behavior to an earlier
"bad experience" (conditioning pathway) because it seemed less
abnormal than to ascribe a phobia to "I saw something
frightening on TV" (vicarious) or "I heard about it"
(informational). Such "social desirability" factors that markedly
bias questionnaire results have been well documented (Crowne &
Marlowe 1964, pp. 3-10).
Retrospective life-event studies in general have been widely
criticized for a variety of potential experimental confounds (Kasl
1983), including subject selection (does the disease or disorder
and/or the act of seeking help affect the recall and the reporting
of life events?), control of events (the degree to which the
adverse events are themselves brought about by the subject's
disorder), timing of events (the onset date of the disorder in
relation to the event), and the subjective, biased, and imprecise
nature of subjects' reported events, particularly for those that are
more distal in time. These and other confounds prompted some
writers more than a decade ago to warn of extreme limitations on
causal inferences in life event studies of mental disorders in
particular (Dohrenwend et al. 1982). More prospective
approaches to the study of adverse life events and their role in the
development of psychiatric and physical disorders have been
strongly urged (Kasl 1983).
The etiology of phobia through classical conditioning
mechanisms may be difficult to confirm in human studies, which
are generally but not necessarily limited by a dependence on
verbal report. Experimental analogues of panic, anxiety, and
phobia have, however, been produced in the animal classical
conditioning and operant laboratory for many decades (see a
review in Lai & Emmett-Oglesby 1983; also Pavlov 1927/1960,
pp. 290-91). Merckelbach & van den Hout note that some
researchers are attempting to reconcile animal and human models
of classically conditioned fear (e.g., Davey 1987). It may be as
important at this stage of the research to develop clean
methodologies that yield unambiguous findings as it is to engage
in additional theory-building.
Tonneau correctly points out the unintentional tug-of-war
between "internal processes" and "functional relationships" as
contradictory approaches to describing classical conditioning
findings in the original target article (Turkkan 1989). The target
article aimed for a functional analysis and attempted to derive
general empirical laws from data resulting from such an approach
(pp. 136-37). The term "associative processes" was used not so
much in the sense of the internal "secrets behind the phenomena"
(Malone 1981, p. 152) but more as an overarching term such as
"learning"; that is, an abstract term nevertheless rooted in
method, and, as Tonneau describes, in correlations between
stimuli. Tonneau also explores transfer of stimulus function in
classical conditioning as a parallel to functional equivalence
effects in operant paradigms. "Transfer of stimulus function" as
defined by Tonneau
396
BEHAVIORAL AND BRAIN SCIENCES (1993) 16:2
appears to be the same learning process conveyed by the earlier
classical conditioning term "stimulus substitution." (I suggest that
"substitution" is a better descriptor than "transfer," which might
almost imply a yielding of control over behavior to another
stimulus.) Some interesting studies using matching-to-sample
designs have shown classical conditioning-like substitution of
stimulus functions from first-order S+ stimuli that are directly
paired with reinforcement to second-order stimuli that are
correlated with the S+ stimuli (e.g., de Rose et al. 1988). Such
studies point up once again the arbitrary assignment of stimulus
properties as "eliciting" or "occasion setting" to classically
conditioned versus operant learning.
NOTE
1. Behavior analysts do not refer to motivational properties as
residing within stimuli, but instead refer to external deprivation
operations.
References
Boelens, H. & Smeets, P. (1990) An analysis of emergent simple discrimination in
children. Quarterly Journal of Experimental Psychology 42B: 135-52. [FT]
Bolles, R. C. (1984) On the status of causal modes. Behavioral and Brain
Sciences 7:482-83. [FT]
(1985) A cognitive, nonassociative view of inhibition. In: Information processing
in animals: Conditioned inhibition, ed. R. R. Miller & N. E. Spear. Erlbaum.
[FT]
Branch, M. N. (1977) On the role of "memory" in the analysis of behavior.
Journal of the Experimental Analysis of Behavior 28:171-79. [FT]
Brown, B. L. (1985) Pavlovian factors in choice behavior. Behavioral and Brain
Sciences 8:333. [IT]
Catania, A. C. (1984) Learning (2nd edition). Prentice-Hall. [FT]
Catania, A. C, Home, P. & Lowe, C. F. (1988) Transfer of function across
members of an equivalence class. Presented at the second European Meeting
on the Experimental Analysis of Behaviour, Liege, Belgium, July. [FT]
Costello, C. G. (1982) Fears and phobias in women: A community study. Journal
of Abnormal Psychology 4:280-89. [HM]
Crowne, D. P. & Marlowe, D. (1964) Approval motive: Studies in evaluative
dependence. Wiley. [rJST]
Davey, G. C. (1987) An integration of human and animal models of Pavlovian
conditioning: Associations, cognitions, and attributions. In: Cognitive
processes and Pavlovian conditioning in humans, ed. G. Davey. Wiley.
[rJST, HM]
(1989) UCS revaluation and conditioning models of acquired fears. Behavior
Research and Therapy 27: 521-28. [HM]
Denny, M. R. (1986) "Retention" of S-R in the midst of the cognitive invasion. In:
Theories of animal memory, ed. D. F. Kendrick, M. R. Rilling & M. R.
Denny. Erlbaum. [rJST]
de Rose, J. C, Mellvane, W. J., Dube, W. V., Galpin, V. C. & Stoddard, L. T.
(1988) Emergent simple discrimination established by indirect relation to
differential consequences. Journal of the Experimental Analysis of Behavior
50:1-20. [rJST, FT]
Dickinson, A. (1980) Contemporary animal learning theory. Cambridge
University Press. [FT]
Dickinson, A. & Mackintosh, N. J. (1978) Classical conditioning in animals.
Annual Review of Psychology 29:587-612. [FT]
Dinsmoor, J. A. (1985) The integrative power of the CS-US interval in other
contexts. Behavioral and Brain Sciences 8:336-37. [FT]
Dohrenwend, B., Pearlin, L., Clayton, P., Riley, M., Hamburg, B. & Rose, R. M.
(1982) Panel report on stress and life events. In: Stress and human health:
Analysis and implications of research, ed. G. R. Elliott & C. Eisdorfer.
Springer. [rJST]
Dube, W. V., Mellvane, W. J., Mackay, H. A. & Stoddard, L. T. (1987) Stimulus
class membership established via stimulus-reinforcer relations. Journal of the
Experimental Analysis of Behavior 47:159-75. [IT]
Eysenck, H. J. (1979) The conditioning model of neurosis. Behavioral and Brain
Sciences 2:155-99. [HM]
Eysenck, M. W. (1988) Anxiety and attention. Anxiety Research 1:9- 16. [HM]
Fantino, E. (1977) Conditioned reinforcement: Choice and information. In:
Handbook of operant behavior, ed. VV. K. Honig & J. E. R. Staddon.
Prentice-Hall. [FT]
Fields, C. (1989) Explaining classical conditioning: Phenomenological unity
conceals mechanistic diversity. Behavioral and Brain Sciences 12:141- 42.
[rri]
Fields. L., Adams, B. J., Verhave, T. & Newman, S. (1990) The effects of
nodality on the formation of equivalence classes. Journal of the
Experimental Analysis of Behavior 53:345-58. [FT]
Fields, L., Verhave, T. & Fath, S. (19*1) Stimulus equivalence and transitive
associations: A methodological analysis. Journal of the Experimental
Analysis of Behavior 42:143-57. [FT]
Furedy, J. J. (1979) Teaching self-regulation of cardiac function through
imaginational Pavlovian and biofeedback conditioning. Remember the
response. In: Biofeedback and self-regulation, ed. N. Birnbaumer & H. D.
Kimmel. Lawrence Erlbaum. [RBM]
(1989)
Flights of Ideological fancy about classical conditioning do not
produce valid science or useful technology. Behavioral and Brain Sciences
12:142-43. [RBM]
Furedy, J. J. & Riley, D. M. (1987) Human Pavlovian autonomic conditioning
and the cognitive paradigm. In: Cognitive processes and Pavlovian
conditioning in humans, ed. G. Davey. Wiley. [RBM]
Furedy, J. J., Shulhan, D. J. & Randall, D. C. (1989) Human Pavlovian HR
decelerative conditioning with negative tilt as US: A review of some S-R,
stimulus-substitution evidence. International Journal of Psychophysiology
7:19-23. [RBM]
Gardner, L. & Malmo, R. B. (1969) Effects of low-level septal stimulation on
escape: Significance for limbic-midbrain interactions in pain. Journal of
Comparative and Physiological Psychology 68:65-73. [RBM]
Catch, M. B. & Osborne, J. G. (1989) Transfer of contextual stimulus function
via equivalence class development. Journal of the Experimental Analysis of
Behavior 51:369-78. [FT]
Goldiamond, I. (1966) Perception, language, and conceptualization rules. In:
Problem solving: Research, method, and theory, ed. B. Kleinmuntz. Wiley.
|IT]
Corniczano. l., Kehoe. E. J. & Marshall, B. S. (1983) Twenty years of classical
conditioning with the rabbit. Progress in Psychobiology and Physiologica
Psychology 10:198-264. [rJST]
Hall, G. (1988) Reasoning and associative learning. Presented at the second
European Meeting on the Experimental Analysis of Behaviour, Liege,
Belgium, July. [FT]
Hayes, S. C. (1989) Nonhumans have not yet shown stimulus equivalence.
Journal of the Experimental Analysis of Behavior 51:385-92. [FT]
Hayes, S. C, Devany, J. M., Kohlenberg, B. S., Brownstein, A. J. & Shelby, J.
(1987) Stimulus equivalence and the symbolic control of behavior. Revista
Mexicana de Analisis de la Conducta 13:361-74. [FT]
Hayes, S. C, Hayes, L. J. & Reese, H. W. (1988) Finding the philosophical core:
A review of Stephen C. Pepper's World hypotheses: A study in evidence.
Journal of the Experimental Analysis of Behavior 50:97-111. [FT]
Hineline, P. N. (1984) Can a statement in cognitive terms be a behavior-analytic
interpretation? Behavior Analyst 7:97-100. [FT]
(1986)
Re-tuning the operant-respondent distinction. In: Analysis and
integration of behavioral units, ed. T. Thompson & M. D. Zeiler. Erlbaum.
[FT]
(1990)
The origins of environment-based psychological theory. Journal of
the Experimental Analysis of Behavior 53:305-20. [FT]
Hitchcock, J. M., Reed, T. L., McCabe, P. M. & Schneiderman, N. (1989) Effect
of lesions of the spinal trigeminal tract on acquisition of classically
conditioned bradycardia in the rabbit. Society for Neuroscience Abstracts
15:888. [RBM]
Holland, P. C. (1981) Acquisition of representation-mediated conditioned food
aversions. Learning and Motivation 12:1-18. [FT]
Holland, P. C. & Forbes, D. T. (1982) Control of conditional discrimination
performance by CS-evoked event representations. Animal Learning 6Behavior 10:249-56. [FT]
Hull, C. L. (1943) Principles of behavior. Appleton-Century-Crofts. [rJST]
Jacobs, W. J. (1989) What is classical conditioning? Behavioral and Brain
Sciences 12:146. [FT]
Jenkins, H. M., Barnes, R. A. & Barrera, F. J. (1981) Why autoshaping depends
on trial spacing. In: Autoshaping and conditioning theory, ed. C. M. Locurto,
H. S. Terrace & J. Gibbon. Academic Press. [FT]
Jenkins, H. M. & Moore, B. R. (1973) The form of the auto-shaped response with
food or water reinforcers. Journal of the Experimental Analysis of Behavior
20:163-81. [FT]
Kapp, B. S., Frysinger, R. C, Gallagher, M. & Haselton, J. R. (1979)
Amygdala central nucleus lesions: Effect on heart rate conditioning in the
rabbit. Physiology and Behavior 23:1109-17. [RBM]
Kasl, S. V. (1983) Pursuing the link between stressful life experiences and
disease: A time for reappraisal. In: Stress research, ed. C. L. Cooper. Wiley.
[rJST]
Kehoe, E. J., Marshall-Goodell, B. & Cormezano, I. (1987) Differential
conditioning of the rabbit's nictitating membrane response to serial
compound stimuli. Journal of Experimental Psychology: Animal Behavior
Processes 13:17-30. [FT]
Keller, F. S. & Schoenfeld, W. N. (1950) Principles of psychology: A systematic
text in the science of behavior. Appleton-Century-Crofts. [FT]
Lai, H. & Emmett-Oglesby, M. W. (1983) Behavioral analogues of anxiety.
Neuropharmacology 22:1423-41. [rJST]
Lazar, R. (1977) Extending sequence-class membership with matching to sample.
Journal of the Experimental Analysis of Behavior 27:381- 92. [FT]
Lazar, R. M. & Kotlarchyk, B. J. (1986) Second-order control of sequence-class
equivalences in children. Behavioural Processes 13:205-15. [FT]
Lee, V. L. (1990) Methodological commitments and other philosophic
necessities. Behavior Analyst 13:91-94. [FT]
Malmo, R. B. (1963) On central and autonomic nervous system mechanisms in
conditioning, learning and performance. Canadian Journal of Psychology
17:1-36. [rJST, RBM]
(1964)Heart rate reactions and locus of stimulation within the septal area of the
rat. Science 144:1029-30. [rJST, RBM]
(1965)Classical and instrumental conditioning with septal stimulation as
reinforcement. Journal of Comparative and Physiological Psychology 60:18. [RBM]
Malone, J. C, Jr. (1981) A fourth approach to the study of learning: Are
"processes" really necessary? Behavioral and Brain Sciences 4:151- 52.
[rJST]
Marks, I. M. (1977) Phobias and obsessions. Clinical phenomena in search of
laboratory models. In: Psychopathology: Experimental models, ed. J. D.
Maser& M. E. P. Seligman. Freeman. [HM]
Marr, M. J. (1983) Memory: Models and metaphors. Psychological Record
33:12-19. [FT]
McNally, R. J. (1987) Preparedness and phobias: A review. Psychological
Bulletin 101:283-303. [HM]
McNally, R. J. & Steketee, G. S. (1985) The etiology and maintenance of severe
animal phobias. Behaviour Research and Therapy 23:431- 35. [HM]
Merckelbach, H., de Ruiter, C, van den Hout, M. A. & Hoekstra, R. (1989a)
Conditioning experiences and phobias. Behavior Research and Therapy
27:658-62. [rJST, HM]
Merckelbach, H., van den Hout, M. A. & van der Molen, G. M. (1989b) The
phylogenetic origin of phobias: A review of the evidence. In: Fresh
perspectives on anxiety disorders, ed. P. M. G. Emmelkamp. Swets &
Zeitlinger. [HM]
Nevin, J. A. (1983) Pavlovian contingencies and conditioned reinforcement.
Behavioral and Brain Sciences 6:711-12. [FT]
Nicholaichuk, T. P., Quesnel, L. J. & Tait, R. W. (1982) Eysenck’s theory of
inculation: An empirical test. Behaviour Research and Therapy 20:329- 38.
[HM]
Pavlov, I. P. (1927/1960) Conditioned reflexes. Dover. [rJST]
Pearce, J. M. & Hall, G. (1980) A model for Pavlovian learning: Variations in he
effectiveness of conditioned but not of unconditioned stimuli. Psychological
Review 87:532-52. [FT]
Rachlin, H., Logue, A. W., Gibbon, J. & Frankel, M. (1986) Cognition and
behavior in studies of choice. Psychological Review 93:33-45. [FT]
Rachman, S. (1977) The conditioning theory of fear-acquisition: A critical
examination. Behaviour Research and Therapy 15:375-87. [HM]
Rashotte, M. E. (1981) Second-order autoshaping: Contributions to the research
and theory of Pavlovian reinforcement by conditioned stimuli. In:
Autoshaping and conditioning theory, ed. C. M. Locurto, H. S. Terrace & J.
Gibbon. Academic Press. [FT]
Reiss, S. (1980) Pavlovian conditioning and human fear: An expectancy model.
Behavior Therapy 11:380-96. [HM]
Rescorla, R. A. (1974) Effects of inflation of the unconditioned stimulus value
following conditioning. Journal of Comparative and Physiological
Psychology 86:101-6. [FT] (1988)
Pavlovian conditioning: It's not what you think it is. American Psychologist
43:151-60. [RBM]
Ross, A. R. & Malmo, R. B. (1979) Cardiovascular responses to rewarding brain
stimulation. Physiology and Behavior 22:1005-13. [RBM]
Schnaitter, R. (1986) Behavior as a function of inner states and outer
circumstances. In: Analysis and integration of behavioral units, ed. T.
Thompson 6r M. D. Zeiler. Erlbaum. [FT]
BEHAVIORAL AND BRAIN SCIENCES (1993) 16:2 397
Seligman, M. E. P. (1971) Phobias and preparedness. Behavior Therapy 2:30720. [HM
Seligman. M. E. P. & Maier, S. F. (1967) Failure to escape traumatic shock.
Journal of Experimental Psychology 74:1-9. [rJST]
Shimp, C. P. (1989) Contemporary behaviorism versus the old behavioral straw
man in Gardner's The mind's new science: A history of the cognitive
revolution. Journal of the Experimental Analysis of Behavior 51:163-71.
[FT]
Sideroff, S., Elster, A. J. & Schneiderman, N. (1972) Cardiovascular classical
conditioning in rabbits (Oryctolagus cuniculus) using appetitive or aversive
hypothalamic stimulation as the US. Journal of Comparative and
Physiological Psychology 81:501-8. [RBM]
Sideroff. S.. Schneiderman, N. & Powell, D. A. (1971) Motivational properties of
septal stimulation as the US in classical conditioning of heart rate in rabbits.
Journal of Comparative and Physiological Psychology 74:1-10. [RBM]
Sidman, M. 6c Tailby, W. (1982) Conditional discrimination vs. matching to
sample: An expansion of the testing paradigm. Journal of the Experimental
Analysis of Behavior 37:5-22. [ FT]
Skinner, B. F. (1977) Why I am not a cognitive psychologist. Behaviorism 5:110. [FT]
Staats, A. W. (1968) Learning, language and cognition. Holt, Rinehart &
Winston. [FT]
Terrace, H. S. (1973) Classical conditioning. In: The study of behavior: Learning,
motivation, emotion, and instinct, ed. J. A. Nevin & G. S. Reynolds. Seott,
Foresman. [rJST]
398
BEHAVIORAL AND BRAIN SCIENCES (1993) 16:2
Tonneau, F. (1990) From reflex to memory: Molar sequences in Pavlovian and
instrumental conditioning. Psychological Record 40:587-607. [FT]
Turkkan, J. S. (1989) Classical conditioning: The new hegemony. Behavioral and
Brain Sciences 12:121-79. [rJST, HM, RBM, FT]
Van Dercar, D., Elster, A. & Schneiderman, N. (1970) Classical heart rate
conditioning motivated by hypothalamic or septal US stimulation. Journal of
Comparative and Physiological Psychology 72:145-52. [RBM]
Wagner, A. R. (1981) SOP: A model of automatic memory processing in animal
behavior. In: Information processing in animals: Memory mechanisms, ed.
N. E. Spear & R. R. Miller. Erlbaum. [FT]
Williams, B. A. (1983) Revising the principle of reinforcement. Behaviorism
11:63-88. [ FT]
(1984) Stimulus control and associative learning. Journal of the Experimental
Analysis of Behavior 42:469-83. [FT]
Wixted, J. T. (1989) Nonhuman short-term memory: A quantitative reanalysis of
selected findings. Journal of the Experimental Analysis of Behavior 52:40926. [FT]
Wixted, J. T. & McDowell, J. J. (1989) Contributions to the functional analysis
of single-trial free recall. Journal of Experimental Psychology: Learning,
Memory, and Cognition 15:685-97. [FT]
Wolpe, J. (1989) The derailment of behavior therapy: A tale of conceptual
misdirection. Journal of Behavior Therapy and Experimental Psychiatry
20:3-15. [HM]
Wolpe, J. & Rachman, S. (1960) Psychoanalytic evidence: A critique based on
Freud's ease of little Hans. Journal of Nervous and Mental Disease 131:13547. [HM]