Attention and para-attentional processing. Event-related brain potentials as tests of a model.

In 1972 when we began to analyze the vast amount of material from the laboratories of physiological psychologists, we had only a vague conceptualization of what a model of attention might look like. We began where everyone else had, with the view that everything had something to do with "arousal" but with Lacey's (1967) warning in mind that all of the dependent variables might not actually be measuring aspects of the same process. With this warning in mind, we were forced by the data to organize them into a three-systems mode. Since the first publication of this model in 1975, we have found increasing amounts of evidence to support and extend it. This evidence is briefly reviewed in the present paper in terms of the techniques employed in various types of investigation. Further, the current review of data has made it possible to specify the para-attentional substrate (the extrinsic lemniscal primary projection systems) upon which the three systems described in the earlier model operate. The earlier model was based on psychophysiological, neurobehavioral and neurochemical analyses while the current specification results from the results of recordings of event-related brain electrical responses. The conclusions derived from these results can be summarized as follows: First. It has become possible to distinguish controlled attention from the para-attentional pre- and post-attentive automatic processes upon which controls operate. Second. The pre- and post-attentive processes appear to be coordinate with activity in the extrinsic lemniscal primary sensory projection systems. Processing in these systems is reflected in the early components of event-related brain electrical potentials. These extrinsic systems are, however, not just throughputs for further processing. Rather, they are sensitive to the history of reinforcement which the subject has experienced. The concept of a limited channel capacity must, therefore, be modified to encompass this ability of organisms to improve, through practice, their competence to process a great deal of information in parallel. Competence, not capacity, limits central processing span. Third. A set of intrinsic extralemniscal processing systems has been identified to operate via a tecto-tegmental pathway to the reticular nucleus of the thalamus. The later components (N2P3, etc.) of event-related potentials have been shown to reflect processing in these systems and those that control them. Activity in these systems has been related to targeted conscious awareness.(ABSTRACT TRUNCATED AT 400 WORDS)