Behao. Res. Thu. Vol. 31, No. 1, pp. 87-95, 1993 0005-7967/93 $5.00 + 0.00 Printed in Great Britain. All rights reserved Copyright 0 1992 Pergamon Press Ltd PREATTENTIVE PROCESSING, PREPAREDNESS AND PHOBIAS: EFFECTS OF INSTRUCTION ON CONDITIONED ELECTRODERMAL RESPONSES TO MASKED AND NON-MASKED FEAR-RELEVANT STIMULI JOAQUIM J. F. SOARES and ARNE OHMAN* Department of Clinical Psychology, University of Uppsala, P.O. 1225, S-751 42 Uppsala, Sweden zyxwvutsrqponmlkjih (Received 13 December 1991) Summary-We hypothetized that autonomic responses conditioned to fear-relevant stimuli, in contrast to responses conditioned to neutral stimuli, can be elicited after only an automatic, non-conscious analysis of the stimulus. Consequently, they may be expected to be insensitive to verbal instructions. Normal subjects were conditioned to either fear-relevant stimuli (snakes or spiders) or neutral stimuli (flowers or mushrooms) in a differential conditioning paradigm with shock as the unconditioned stimulus. In a subsequent extinction series, half of the subjects were shown the conditioned stimuli under masking conditions preventing their conscious recognition, whereas the other half were exposed to non-masked stimuli. Then half of the subjects in each of the masking conditions were verbally instructed that no more shocks would be delivered and then the extinction trials followed. Consistent with our hypothesis, differential responses to the fear-relevant CSsf and CSs- remained unaffected by both masking and instruction, whereas differential responding to neutral stimuli was wiped out by the masking procedure and the verbal instruction. INTRODUCTION Substantial evidence attest to the selectivity of human fears, that is, to the tendency of some stimuli to elicit fear more easily than others (e.g. Agras, Sylvester & Oliveau, 1969; Marks, 1969). Seligman (1970, 1971; Seligman & Hager, 1972) suggested that this specificity of fear indicates a biologically originated preparedness to associate fear more easily with objects representing a potential threat than with neutral objects. Thus, the preparedness model (Seligman, 1971) was based on the observation that most simple clinical phobias involve potentially threatening stimuli and situations (de Silva, Rachman & Seligman, 1977) of vital importance to the survival of mankind. Seligman (1971) argued that as prepared responses phobias are selective with respect to stimulus events, quickly acquired and difficult to extinguish, and that they are insensitive to cognitive manipulations. The preparedness notion was empirically tested by ijhman and co-workers in a series of studies on human conditioning to fear-relevant stimuli, e.g. snakes, spiders and angry faces (see McNally, 1987; ahman, 1992b; for reviews). It was demonstrated that electrodermal responses previously conditioned to fear-relevant stimuli by an aversive unconditioned stimulus (UCS) were slower to extinguish than responses conditioned to neutral stimuli (Fredrikson & ijhman, 1979; ijhman & Dimberg, 1978; ohman, Fredrikson & Hugdahl, 1978; ohman, Fredrikson, Hugdahl & Rimmii, 1976). This effect was not dependent on a real presentation of the UCS, because threatening instructions and observational learning were as effective as direct conditioning in producing enhanced resistance to extinction to fear-relevant but not to neutral stimuli (Hugdahl, 1978; Hugdahl & ijhman, 1977; Hygge & ohman, 1978). Finally, Hugdahl (1978) and Hugdahl and i)hman (1977) reported that verbal instructions had different effects on the extinction of conditioned electrodermal responses to fear-relevant and neutral stimuli. Responses to fear-relevant stimuli were not influenced by instructions eliminating the expectancy of future UCSs, while responses to neutral stimuli were abolished by this experimental manipulation. This dissociation between expectancy and autonomic responding suggests that the conditioned responses (CRs) to fear-relevant stimuli may have been *Author for correspondence. 87