Derived generalization of attentional bias for laboratory-induced threat: Yes but.

There is laboratory evidence that fear conditioning underlies the emergence of attentional bias (AB) for threat. Our main objective was to test, for the first time, whether derived or symbolic responding contributes to the generalization of AB across non-conditioned stimuli. Participants were all university students ( = 86) with no pre-existing conditions. We first employed an exogenous cueing paradigm with two color slides (i.e., A1 or to-be CS+, and A2 or to-be CS-) serving as cues, and loud white noise serving as unconditioned stimulus during conditioning trials. We then employed a match-to-sample procedure to establish a derived equivalence relation between color A1 and arbitrary shape C1 as well as between color A2 and arbitrary shape C2. Next, we investigated the transfer of AB across non-conditioned stimuli: participants performed the same spatial cueing task with non-conditioned C1 and C2 stimuli serving as cues. Results replicated previous findings on the conditioning basis of AB, and most importantly, showed preliminary evidence of AB transfer: those participants who appraised C1 and not C2 as a signal of impending noise showed AB toward C1. This is the first laboratory demonstration that AB may generalize to stimuli physically unrelated to directly conditioned threats. Unfortunately, the small number of participants showing this effect calls for cautious considerations.