Search This Blog

Sunday, October 24, 2004

Incest and Learning: A Bio-Behavioral Explanation

The aversion to incest is one of the most pervasive and remarkable regularities of human behavior, yet to this date has defied explanation. The fact that early and prolonged exposure to opposite sex members of kith and kin results in sexual apathy has been matched with hypotheses that explain incest aversion, but have no true explanatory or predictive power. That is, incest aversion is normally adduced to a specialized cognitive module or 'imprinting mechanism' (Pinker, 1997; Tooby and Cosmides, 1995; Wilson, 1998), yet such an explanation does not account for the latency, generality, or mutability of the response, nor does it match any actual processes, real or hypothesized, that are instantiated in the human brain. Although it may be argued that a specialized mental or neural module is as necessary or demonstrable as phlogiston (the imaginary component that enabled fire), it can only be abandoned if general purpose neural processes are demonstrated to be recruited in early experience to produce the same results.

The author proposes that such a mechanism does exist, but has been ignored not because of its complexity or obtuseness to observation, but rather because of differing meta-theoretical principles regarding the informing principles of psychological science. The specific question is whether cognitive structures that underscore incest aversion are informed from the teleological (i.e. imputed design in nature) principles of evolutionary psychology or the experimental and ethologically (i.e. animal experiments) informed principles of behaviorism.

Incest and Evolutionary Psychology

The environmental circumstances that frame incest aversion are clear. Prior to puberty, a history of continuous social contact with members of the opposite sex, regardless of their biological relatedness to each other will result in a generalized reduction or elimination of the preparatory response of sexual arousal. According to the teleological principles that guide evolutionary explanations of cognitive phenomena, the lack of sexual arousal resulting from early childhood social contact is due to metaphorical 'imprinting' mechanisms that are inferentially derived from hypothetical selection pressures occurring over the evolutionary span of human development. In other words, because interbreeding between closely related kin results in the likely physiological impairment of their offspring, an imprinting mechanism evolved, the existence of which was predicated to ensure the long term 'fittedness' of the race.

Although the correlations between environmental events such as proximal rearing of different sex individuals and incest aversion is well established, the gradual development of this trait only partially fits the empirical data. Indeed, incest aversion can be displayed towards kith and kin who have not grown up in proximity with one another, and the onset of sexual apathy or aversion can be near instantaneous depending upon changes in information alone (Harris, 1989). Thus we may be automatically averse to incest with in-laws, adopted children, first cousins, etc. with the independent variable being not rearing proximity, but cultural mores or norms that are independent of proximity.

Besides representing incomplete explanations for incest aversion, evolutionary mechanisms presently represent unrealistic explanations. That is, for explanations that implicate modular evolved imprinting mechanisms in lieu of ascertaining real learning processes and the physiological events that comprise them, such processes are wholly inferred from evolutionary selection processes that are themselves inferred. These modular imprinting mechanisms derive from metaphorical perspectives of a human brain that possesses numerous hard-wired computational structures or modules that determine the function and shape of specific behavioral traits. Moreover, selection pressures inferred from the geological record determine the content and structure of these processes. Because of the implied certainty of this presumption, inferred biological processes co-opt the need to discover actual learning and biological processes because the brain acts in an orderly computational manner that is set in motion by hard wired circuits selected by evolution, thus obviating the need for seriously considering a bio-behavioral perspective. In this manner, the metaphors of computation replace the actual processes of cognition or learning that drive human brains and human behavior.

In order to arrive unchallenged to this conclusion, a learning theoretical or behaviorist approach to the brain must be demonstrated to absolutely adhere to the metaphor of the brain as a computational organ bereft of nativistic influences, and thus be dismissed a-priori as an alternative source for explanations for incest aversion, and indeed for the many behavioral patterns (e.g. aggressiveness, altruism, etc.) explained by modular hard-wired cognitive mechanics of evolutionary psychology. In principle, this equates behaviorism with the 'Standard Social Science Model' (Cosmides, Tooby, and Barkow, 1992) that holds that the mind is devoid of any nativistic or inborn influences and is metaphorically akin to a 'blank slate'.

Behaviorism thus became a non-starter as an investigative tool to uncover the covert and instinctive mechanisms equipped by evolution in the human mind. It was, in the words of the evolutionary psychologist Steven Pinker, 'dead' (1997). The idea that behaviorism is moribund or dead reflects the popular view that behavioristic learning theories are purely computational, and conceptualize the human mind as a blank slate that is engraved by experience alone. But at root, behavioristic learning theories have never been exclusively informed by simplistic data processing languages, and modern behaviorism in particular is committed to the more catholic view that behavior is modulated by multiple processes that are primarily non-computational, originate in different areas of the brain, and can be described by multiple sets of metaphors, from Skinnerian contingencies and Pavlovian associational bonds to cognitive metaphors such as expectancy and affect. Thus may be best illustrated by a behavioristic account of incest aversion. .

Incest and Behaviorism

In evolutionary psychology teleological viewpoints are fundamental, and physiological viewpoints incidental. The opposite is the case with biological learning theory or bio-behaviorism. The difference is that unlike evolutionary psychology, a bio-behaviorism provides the opportunity for empirical test, and generalization to similar behaviors from established principles that are rooted in empirical data. Ironically, modern biologically informed behaviorism or bio-behaviorism does not adhere to 'blank slate' information processing metaphors, but entails the unique participation of neural processes that are analog, not digital in nature. In particular, the instinctive or nativistic behaviors described by bio-behaviorism are not neurally embodied as self contained 'modules', but rather derive from the interoperation of different well-defined morphological structures in the human brain.

To understand how learning principles may generalize across types of behavior, and in particular to incest aversion, the concept of 'blocking' or 'latent inhibition' provides a fitting example. Blocking demonstrates not how new information may override old, a common theme implied by blank slate metaphors, but how old information may override new, and thus provide a learning equivalent to instinctive mechanisms that posit how old information encoded by the genome drives instinctive traits. The concept of blocking (Kamin, 1969) entails that the learning associated with one stimulus may inhibit or 'block' any additional associations with that stimulus. For example, present an auditory tone repeatedly for an experimental animal, and the subsequent pairing of the tone with food will result in slower acquisition of the conditioned response, and in some cases no conditioning at all. In other words, the animal will not anticipate food because previous associations of the tone with no food block the associative strength of the tone and food. Moreover, the level or prior training affects the level of associative strength gained on each training trial. Thus, an animal that undergoes multiple trials of the tone without food will be less likely to associate the tone with food than an animal that is subjected to a few trials.

Recent iterations of blocking experiments with humans and animals have demonstrated that associations are not blocked, but rather that inattention to logically salient events is learned that inhibits the behavior those associations may logically entail (Kruschke & Blair, 1999). This concept of learned inattention implies that for behavior to occur, attention to or anticipation of a salient event is as important as the perceived logical or contingent relationship between an event and behavior, and is equally and separately described by learning principles. For the neuropsychology of incentive motivation, the bifurcation of motivation into two separate processes of anticipation and reward, or of wanting and liking (Berridge, 2001) demonstrates that the contingent logic of cause and effect or behavior and reward is not enough to elicit behavior. Attention must be invested, and attention must be learned. This importance is due to the fact that an attentive or anticipatory cognitive state is not a neutral but an affective state that is different from consummatory behavior (e.g. eating, sex) and reflects different neurological processes (namely the activity of midbrain dopamine neurons) that activate global areas of the brain and give value or incentive salience to behavior (i.e. it feels good). Moreover, these processes in turn can be influenced or modulated by experience or learning. Thus, looking forward to or anticipating events (i.e. wanting) represents an entirely different neurological process from having (i.e. liking), and can be altered by learning. Extending this concept to sexual desire, the 'wanting' part of arousal is entirely different physiologically and psychologically from it's 'having' or coital element, and is subject to the often nonconscious vagaries of learning. The major question is whether this observation of 'learned inattention' may transfer without loss to the preparatory response of sexual arousal, and thus account for incest aversion. Continuous exposure to opposite sex members where sexual behavior cannot occur (e.g. as in pre-puberty) and will not be anticipated to occur (e.g. sexual modeling or fantasy) will result in a continuous association of a stimulus event (the girl next door) with the absence of a sexual priming response or attentive arousal. Thus inattention or unanticipation is learned, rather than a contingency (boy meets girl) being unlearned. In other words, the contingency between behavior and 'getting the girl' is not sufficient itself to generate behavior. The individual must learn to be attentive to a lady's charms, and if he has learned to be inattentive, he will not respond. Secondly, anticipation as a covert operation or behavior can have negative implications that act separately or in concert with blocking mechanisms. Thus, we will not think or fantasize about having sex with close kin because it strongly implicates social norms, and if those norms are suddenly imposed, anticipation stops, and with it the 'drive' to have sex with kin.

In sum, sexual anticipation is different psychologically and physiologically from sexual attraction. One can find a sibling attractive, but not anticipate sex or be inclined to anticipate sex. Anticipation, or the covert modeling of sexual behavior, can be influenced by learning (negative reinforcement of societal norms) or the interference of prior learning (blocking). Thus behaviorism can account for all aspects of incest aversion, and demonstrate the behavior as fully representative of learning processes. In other words, sexual anticipation as an affective state represents the activity of midbrain dopamine neurons that receive input from the neocortex, and are thus constrained by learning principles.that are informed by the facts of behavior and the facts of the human brain. Thus, one can be aroused by a sexual opportunity because of information about the opportunity, and likewise be unaroused when an opportunity is unavailable, entails negative outcomes (e.g.,strong social disapproval) or if an individual is physically incapable (due to physical immaturity) of being aroused.

One question, as of yet unanswered, is whether a core principle of learning theory such as blocking is a valid explanation for incest aversion, and if it can be explained by the data language of Pavlovan behaviorism and the neurological events that language underscores. This will require the experimental manipulation of sexual anticipation, an element of desire that has not been controlled for in any representative studies on incest aversion. In lieu of an experimental verification of this hypothesis, a Pavlovian explanation of incest aversion certainly fits the data better than the 'critical period' process of sexual imprinting. It can explain how we can be resistant to sudden rises in sexual temptation, and also how sexual aversion can automatically happen, as in the literary example of Oepidus or in more common situations when adults become biological or adoptive parents parents and 'automatically' become incest averse. But as importantly, it can explain incest aversion or apathy without the postulation of additional neurological or cognitive baggage, and demonstrate the economy of means employed by the human brain in generating behavior of staggering diversity.

Behaviorism, Evolutionary Psychology, and Incestuous Science

A behavioristic explanation of incest is hypothetical, yet testable, and it is the ability to be subject to experimental falsification that distinguishes it from evolutionary explanations that are untestable, and hence cannot meet the strictures of science. But the continuing neglect and disparagement of behaviorism, and by implication, of learning theory is due in large measure to a gross misunderstanding as to what behavioristic learning theories entail.

The purpose of the scientific enterprise is to consider all viable and testable explanations for phenomena, but the neglect of alternatives can be rationalized by prejudicial factors that unwarranted by the evidence. This is particularly the case regarding the nature of behavioristic learning theories. Learning theories describe how experience influences behavior, and come in many stripes, from Freudian constructs to mathematical models. As a special class of learning theory, a 'behaviorism' uses ethological data, is deductive in nature (although Skinnerian methodological and radical behaviorisms are a notable exception), employs within (single subject, multiple trials) group designs to isolate and study psychological variables, considers behavior in all its manifestations (overt, covert, molar, molecular), and because it uses animal behavior to inform an understanding of human behavior, limits the metaphorical representations of the data it collects, hence the nom de guerre 'behaviorism'. Behavioristic research is not informed by teleological reasons that assign a purpose to behavior beyond what the data would permit. On the other hand, evolutionary psychology is highly teleological in nature, and presumes that stories that impute an evolutionary design can provide a precise specification of the functions of cognitive mechanisms (Tooby and Cosmides, 1995). Nonetheless, teleological stories that imply an ultimate purpose to behavior, namely insuring reproductive fitness, never inform behavioristic research in any of its guises, from the study of overt molar behavior (Thorndike, Pavlov, Skinner, Tolman) to covert neural behavior (Donahoe, Berridge), because teleological causes are at root untestable, and conflict with the Popperian principles of falsification or testability that inform the biological and physical sciences. Thus, evolutionary psychology and behavioristic learning theory are incompatible because of their opposing estimates of the heuristic or research value of teleological versus ethologically guided experimental principles, and more significantly, their differing estimates as to the definition of scientific inquiry. That behaviorism and evolutionary psychology do not inform and rather neglect each other is rooted in the denial of the epistemological principles that support each subject matter. and can only change if one or the other abandons the very definitions of their fields. That is, if teleological principles are not needed to explain incest, then they are likely not needed for other behavioral traits that are equally amenable to a bio-behavioristic explanation. But if teleology is abandoned as an essential informing principle in science, then evolutionary psychology is refute at root, and must be replace ironically by the very behaviorism it has trumpted as dead.

Does an explanation for incest need evolutionary psychology?

A case in point is incest aversion. The Kiplingesque stores that impute final causes to incest aversion cannot change in any manner how incest avoidance is observed to develop in naturalistic settings. It suggests neither procedure not process, yet by its premature closure on the matter, can preclude the use of alternative and testable hypotheses that describe its true nature. Indeed, it may be persuasively argued that fitness issues have nothing whatsoever to do with incest, and that incest avoidance is, like a spandrel, a by-product of unremarkable learning processes that are traceable in human behavior and in the activity of the human brain. Unfortunately, whereas behaviorism neglects evolutionary psychology, the reverse holds true as well. Indeed, evolutionary stories rarely if ever are informed by ethological or neural principles (Panksepp, 2000). To describe how the mind works without referencing the biological mind, a claim characteristic of evolutionary psychology and dogmatically made by the evolutionary psychologist Steven Pinker (1997) in his ironically named book 'How the Mind Works', highlights these very opposing positions.

To define psychological science through the prism of philosophical rather than empirical perspectives limits psychology to an incestuous brand of science, wherein quoting from members who hold the same tribal affiliation is the way for progress. This results in endless debates on the workings of the world. The fact that modern behavioral psychologists can mix and match the once exclusive data languages of Skinner, Pavlov, and cognitive science, and are grounded to an understanding of the organic brain is a hopeful sign. But whether psychology can match the progress of the physical and biological sciences in understanding and controlling our world and ourselves is dependent upon a renewed dedication to scientific principles, and how our behavior and consciousness itself emerges from the simplest processes in brain and body that are demonstrated to us from the behaviorisms that so many academics view as dead.


Berridge, K. (2001) Reward Learning: Reinforcement, Incentives, and Expectancies,
The Psychology of Learning and Motivation, (3), Academic Press, New York

Harris, M. (1989) Our Kind, Harper Collins: New York

Kamin, L. (1969) Predictability, surprise, attention, and conditioning. In Campbell, B. and church, R., eds. Punishment and aversive behavior. New York: Appleton Century-Crofts; 44, 276-296

Kruschke J. K. and N. J. Blair (2000) Blocking and backward blocking involve learned inattention, Psychonomic Bulletin and Review, 7(4), 636-645

Panksepp, J. and J. B. Panksepp (2000) The Seven sins of evolutionary psychology. Evolution and Cognitivion, 6(2), 108-131

Pinker, S. (1997) How the Mind Works. Norton: New York

Tooby, J. and L. Cosmides (1995) Mapping the Evolved Functional Organization of the Mind and Brain. In Michael S. Gazzaniga (ed.), The Cognitive Neurosciences (1185-1197). Cambridge, Ma: MIT Press

Wilson, Edward O. (1998) Consilience: The Unity of Knowledge. New York: Alfred A. Knopf

No comments: