Promise versus reality in relation to the unitary orienting reflex: A case study examining the role of theory in psychophysiology

https://doi.org/10.1016/j.ijpsycho.2006.01.004Get rights and content

Abstract

We commonly teach beginning science students that theory generates hypotheses which direct our research, framing our experimental observations; and that in turn, these supply the data which support or contradict theory, allowing its self-correction and further development. These propositions are explored here in the context of psychophysiology, concentrating on examples in relation to the Orienting Reflex (OR). It is demonstrated that the realist approach generally portrayed in our teaching about theory and theory testing is not the dominant ethos in this field. Indeed, we pay little more than lip-service to the ideal we teach about. One outcome described here is that the promise offered us by the OR in the 1960s has not been realised. It is argued that this failure may be symptomatic of much of science, suggesting that we need to consciously work to raise the perceived value of theory. The ideation element of theoretical activity can be fostered, and it is argued that we need to do so explicitly at a number of levels in the education of our students, beginning in the early coursework stages and continuing in their research training and development.

Introduction

Few would argue against the premise that science advances only through the constant interplay of theory and observation. This is what we commonly teach beginning science students under the heading of “the scientific method”. We teach them that theory generates hypotheses, that these hypotheses direct our research, and frame our experimental observations. In turn, these observations supply the raw data which can be statistically tested to support or contradict the theory. Through this process, theory is said to be self-correcting, and in continuous development. But it is clear from a cursory examination of any science journal that this process is not operating, or at least is not made explicit, in many studies reported in the literature. Rather, many papers appear to be focussed only on the generation of information/knowledge — the gathering of descriptive data about some phenomenon, primarily of interest to the authors. If the phenomenon is novel and appeals to a wider audience, there may be claims and counter-claims about that description, until some consensus eventually emerges in the literature. All this may occur without any elements of theory being mentioned, let alone seriously addressed. When theoretical elements do emerge, they often appear as a summary device for the description of the phenomenon – a simple generalisation or more-complex model – rather than the full hypothesis-generating theory we teach about. This suggests that our science operates in cycles containing discrete stages of data gathering, data confirmation, modelling, further data gathering, and later integration of models into theories, rather than a continuous interplay between theory and observation (although that stylised interplay may occur at particular points in the cycle). Further, the movement from stage to stage in the cycle seems almost random rather than systematic in nature, raising questions about how progression is triggered. Can the shift from observation to model building, etc., be facilitated to move science forward more efficiently? Such questions touch on the sociology and politics of science, usually ignored in our current “education” of scientists. Some of these propositions and questions are beyond the scope of the current paper, but the core issues are explored here in the context of psychophysiology, concentrating on examples in relation to the Orienting Reflex (OR).

Throughout my research life in psychophysiology, I have attempted to work within the realist approach to science, valuing a particular theory in terms of its supportive evidence and predictions, rather than seeing it merely as a convenient vehicle for discussion. The importance of this distinction for psychophysiology, and particularly for the development of OR theories, was argued a quarter of a century ago by Furedy and Arabian (1979). They contrasted the realist view (Popper, 1959) – theory must pay due regard to the data it supposedly encompasses – with the so-called instrumentalist view (Kuhn, 1962) — theory can be evaluated in terms of its influence or heuristic value. These are philosophical perspectives, alternatives which the scientist is free to choose as part of their personal approach to the profession, but it is not my intention here to argue on philosophical grounds. Rather, I will discuss the implications and issues arising from a realist approach, and point to problems which are apparent as a result of what can be described as an all-too-common adoption of the instrumentalist view.

Science aims to gather and systematise data, i.e., accumulate observations of the universe. Within science, modern psychology largely restricts itself to organismic behaviour (in broad terms encompassing thoughts and emotions) and, to the extent that it has an influence on such behaviour, our environment. These observations/data are the units of our psychological science, but science (as distinct from art) is necessarily characterised by a constant movement beyond the empirical, deriving inferences and building these into coherent groupings. Although based on observation, theory goes beyond the data — it integrates and structures the data into statements of the functional relationships between variables. This theoretical element, the addition of the theorist's ideas to the empirical dataset, can range from a simple logical ordering of the observations into a generalisation, to the integration of those generalisations into a set of relatively well-articulated abstract hypotheses — a theory.

That is, theory generation and development involves the addition of ideas/inferences to a well-structured or coherent dataset. The aim of this activity is to produce a systematisation which describes the phenomenon of interest more economically than the data listing – i.e., it efficiently summarises that dataset – and helps our understanding/conceptualisation of that phenomenon. It does this by facilitating extensions beyond the dataset, i.e., by allowing the generation of novel predictions. The realist looks to testable predictions as the basis for valuing the theory, and if the predictions are not supported by new data, insists that the theory be modified to accommodate the new data. If it cannot be modified and extended to incorporate the previous and new data, then it should be rejected. This is a critical difference between the realist and instrumentalist approaches. Rather than rejecting a falsified theory, the instrumentalist may retain it for its usefulness in talking about or “understanding” the data which it does accommodate, ignoring the failure to predict/describe/accommodate other data. I have placed “understanding” in the previous sentence in quotation marks because I do not consider that the instrumentalist genuinely grapples with the theoretical elements used in this fashion, as it requires some ignoring of the truth of the situation in favour of its convenience. To the realist, the instrumentalist approach is unacceptable, as it is seen to allow the continued existence of flawed theory, contrary to the seeking of truth which underlies real science.

I will endeavour to demonstrate below that the present ignoring of the importance of the OR within psychophysiology arises (at least partly) from the acceptance of flawed theory — or its near-cousin, the atheoretical approach, which avoids theory altogether. This demonstration will centre on aspects of the development of Preliminary Process Theory (PPT) (Barry, 1984a, Barry, 1987a, Barry, 1987b, Barry, 1996). I am using this particular vehicle, not because it should be presented as a superior example of realist theorising, but rather, because of the insights its development has afforded me in relation to the sociology and politics of science, which I think have tended to hinder or pervert the rational development of our science.

Over the years there have been numerous lists of criteria for the evaluation of theory proposed across many disciplines. I have alluded above to the major differences between Popper and Kuhn, but apart from such basic disagreements, many criteria lists have useful elements in common. For example, Dennis and Kintsch (in press) include descriptive adequacy, prediction and falsifiability, parsimony, precision and interpretability, coherence and consistency, postdiction and explanation, originality, breadth, usability, and rationality. Such evaluative criteria should be kept in mind as theoretical developments are described throughout this paper.

Section snippets

Development of preliminary process theory of OR elicitation and habituation

This section initially sketches the empirical and theoretical context of an important focus of psychophysiology in the 1960s and 1970s, largely provided by Sokolov's description of the physiological complex of the OR. This is followed by an outline of my early empirical work in that context, my thinking about those data which led to disagreements with the mainstream perspective, and consequent theoretical developments in relation to an alternative OR theory. In this presentation, I will

Reflections on the interplay between research and theory

I have sketched above a personal perspective on theory development, illustrating the constant interplay between data and ideas/theory through some 35 years of research. This can be used to framework the following comments on theory in psychophysiology.

Conclusions

This paper has aimed to demonstrate the importance of theory in psychophysiology, by pointing to some of the lost opportunities resulting from a lack of focus on theory, its testing and development. From my perspective, the major loss is that the promise of the OR as a widely relevant major organising construct has not been realised. I am sure that workers in other areas could, on reflection, point to similar losses which can be attributed to the current casual attitude to the role of theory in

References (58)

  • R.J. Barry et al.

    EEG alpha activity and the ERP to target stimuli in an auditory oddball paradigm

    International Journal of Psychophysiology

    (2000)
  • R.J. Barry et al.

    Preferred EEG brain states at stimulus onset in a fixed interstimulus interval auditory oddball task, and their effects on ERP components

    International Journal of Psychophysiology

    (2003)
  • R.J. Barry et al.

    EEG differences in children as a function of resting-state arousal level

    Clinical Neurophysiology

    (2004)
  • R.J. Barry et al.

    Event-related potentials in the auditory oddball as a function of EEG alpha phase at stimulus onset

    Clinical Neurophysiology

    (2004)
  • R.J. Barry et al.

    Caffeine effects on resting-state arousal

    Clinical Neurophysiology

    (2005)
  • T.W. Budd et al.

    Decrement of the N1 auditory event-related potential with stimulus repetition: habituation versus refractoriness

    International Journal of Psychophysiology

    (1998)
  • J.A. Rushby et al.

    Separation of the components of the late positive complex in an ERP dishabituation paradigm

    Clinical Neurophysiology

    (2005)
  • P. Tremayne et al.

    Elite pistol shooters: physiological patterning of best versus worst shots

    International Journal of Psychophysiology

    (2001)
  • M. Velden et al.

    Time-dependent bradycardia: a new effect?

    International Journal of Psychophysiology

    (1987)
  • R.J. Barry

    Failure to find evidence of the unitary OR concept with indifferent low-intensity auditory stimuli

    Physiological Psychology

    (1977)
  • R.J. Barry

    The effect of “significance” upon indices of Sokolov's orienting response: a new conceptualisation to replace the OR

    Physiological Psychology

    (1977)
  • R.J. Barry

    Physiological changes in a reaction-time task: further problems with Sokolov's dimension of stimulus “significance”

    Physiological Psychology

    (1978)
  • R.J. Barry

    Novelty and significance effects in the fractionation of phasic OR measures: a synthesis with traditional OR theory

    Psychophysiology

    (1982)
  • R.J. Barry

    Primary bradycardia and the evoked cardiac response in the OR context

    Physiological Psychology

    (1983)
  • R.J. Barry

    Stimulus omission and the orienting response

    Psychophysiology

    (1984)
  • R.J. Barry

    The evoked cardiac response under processing load

    Physiological Psychology

    (1984)
  • R.J. Barry

    Trials effects in the evoked cardiac response under processing load

    Physiological Psychology

    (1984)
  • R.J. Barry

    Heart rate deceleration to innocuous stimuli: an index of the orienting response or stimulus registration?

    Physiological Psychology

    (1986)
  • R.J. Barry

    Preliminary processes in orienting response elicitation

  • Cited by (59)

    • Concealed Information Test

      2018, Detecting Concealed Information and Deception: Recent Developments
    View all citing articles on Scopus
    View full text