Go-no go avoidance discriminations in rats with simple “go” and compound “no go” signals: Stimulus modality and stimulus intensity

Abstract

Four groups of 16 rats were trained in a go-no go discrimination with symmetrical negative reinforcement (active-passive avoidance task). A 2 by 2 design was used with stimulus combinations as one factor [light-go, noise/light-no go, (L + NL -) vs noise-go, light/noise-no go (N+ LN-) J and noise intensity as the other factor (90 dB, high, vs 70 dB, low). The two N+ LN- groups learned both active and passive avoidance responses much more rapidly than the L+ NL - groups. In later phases of training, high noise intensity facilitated passive avoidance in the L+ NL - condition and exerted a slight effect in the opposite direction in the other condition. While the differences in active avoidance learning were the same as in previous work on light and noise C5s, the interactions between stimuli of different modalities appeared to be more important than stimulus modality or stimulus intensity per se in determining rate of passive avoidance learning. Seve ral studies have emphasized the role of conditioned stimulus (CS) factors in the acquisition of instrumental avoidance responses. It is well known that. for rats, noise is a more efficient active avoidance (AA) signal than light (Smith, McFarland, & Taylor, 196L Biederman, 1967), but it must also be considered that it produces stronger unconditioned motor responses and more pseudoconditioning phenomena (see discussion in Morgan, 1968). Recent data on go-no go avoidance discriminations (active-passive avoidance tests) have provided new evidence about the effects of acoustic and visual (Ss and of their interactions (Rosie, Frontali, & Bignami, 1969; Rosie & Bignami, 1970). Rats trained with noise as go (AA) signal and light plus noise as compound no go (passive avoidance, PA) signal (Nt LW-) learned much more rapidly than animals trained with the opposite arrangement (Lt NL-). The greater difficulty of the latter discrimination was due not only to a retardation of AA learning, as expected on the basis of the above data, but also to an impairment of PA learning. The latter difference could not be accounted for either by sensory factors or by differences in the response modulating properties of light and noise per se. In fact, both discriminations with simple go and no go signals (Lt N-, N L-) were learned with few PA er.ors. This indicates that the differences were due to an interaction between the two components of the no go signal, which cannot be explained on the basis of the available behavioral and neurophysiological evidence. The literature data on the interactions between stimuli of different modalities (effects on the startle reflex: Ison & Hammond, 1971; effects on classical conditioning: e.g., Kamin, 1969; Rescorla, 1969; Sutherland & Mackintosh, 1971) suggest that the differences mentioned above could be due to a variety of factors. An obvious problem to be solved prior to further investigation was whether the interactions between noise and light stimuli, leading to the diffe re nce s between the Nt DI- and U ~L­ discriminations, were really due to stimulus modality rather than to stimulus intensity effects. In the present experiment, the acquisition of Nt LN- and Lt NL­ discriminations was studied with two different noise intensities, using the same light intensity as in the work quoted above. In order to have a bet ter control of intensity level, a speech noise from a standard generator was preferred to the buzzer used previously. This was based on a preliminary study, not to be reported in detail, showing that the L+ NL- discrimination was much more difficult than the N LN- discrimination, both with the buzzer noise and with the speech noise. at approximately similar intensities (90 dB; see Method). Furthermore, as in one of the previous experiments (Rosie & Bignarni, 1969. Experiment 4), rats were not given AA pretraining, so as to avoid possible effects of primacy of AA training and interactions between this and other factors.