Temporal encoding as a determinant of inhibitory control

A Pavlovian conditioning experiment explored the possibility that signals for unconditioned stimulus (US) omission provide information about the temporal location of the omitted US. Training with conditioned stimuli (CSs) and a footshock US established multiple signals for US presence (A and B, called training exciters) and multiple signals for US omission (X and Y, called inhibitors). Rats were exposed to two reinforced and two nonreinforced trial types within sessions. Trial types for Group Serial were A --> +, B+, XA-, and Y --> A- and for Group Simultaneous were A --> +, B+, XB-, and Y --> B- (in which A was serially [A --> +] and B was simultaneously [B+] conditioned with the US, X served as a simultaneous inhibitor [XA- or XB-] and Y served as a serial inhibitor [Y --> A- or Y --> B-]). Thus, each group was trained with a simultaneous and serial inhibitor but differed in whether the training exciter whose US was omitted on nonreinforced trials was serially or simultaneously conditioned with the US. A summation test with an independent transfer exciter revealed that inhibitory control was maximal for the simultaneous inhibitor (X) when the relevant training exciter was serially paired with the US (Group Serial) but was maximal for the serial inhibitor (Y) when the relevant training exciter was simultaneously paired with the US (Group Simultaneous). The summation data were predicted by a theoretical account in which inhibitory control is predicted to be maximal when the time at which the US is expected to be presented (information activated by the transfer excitor) is concordant with the time at which the US is expected to be omitted (information activated by the inhibitor). The results suggest that signals for US omission provide information about the temporal location of the omitted US. (C) 1996 Academic Press, Inc.