Effects of signals preceding and following shock on baseline responding during a conditioned-suppression procedure.
The order of warning and aversive decides whether you get brief suppression, later speed-up, or lasting baseline loss.
01Research in Context
What this study did
Davis et al. (1976) tested three ways to pair a warning tone with mild foot-shock in rats. The animals first worked on a steady lever-press schedule for food. Then the team added either tone-shock, shock-tone, or tone-shock-tone. They watched how each order changed the rats’ everyday response rate.
What they found
Tone-shock cut responding and dragged the baseline down. Shock-tone lifted pressing during the tone but still lowered the overall rate. Tone-shock-tone first suppressed the first tone, left the second tone untouched, and kept the baseline closest to normal. Order, not just presence, steered the outcome.
How this fits with other research
Hamilton et al. (1978) later extended the idea to wheel-turn avoidance. They saw the same brief dip then speed-up with foot-shock, but tail-shock stayed purely suppressive. The two studies line up: foot-shock can excite after a tiny pause, while other shock types cannot.
Macdonald et al. (1973) conceptually replicate the benefit of any signal. They showed a single warning tone hurts less than silence before shock, matching H et al.’s finding that even a partly suppressive sequence still beats no signal at all.
Blackman (1970) seems to clash at first glance. That paper reported a pre-shock tone wrecked temporal discrimination. H et al. found the same tone order kept the baseline closest to normal. The difference is the task: D required rats to time two responses in a chain, while H used a simple steady lever press. Complex timing suffers more than plain rate, so the papers disagree only on task difficulty, not on the value of a warning.
Why it matters
If you must embed unavoidable aversives in a program, first decide what you need to protect. A simple keep-going response survives best with a signal-shock-signal wrap-around. A delicate response chain may still crumble even with the same wrap. Test the exact sequence, not just ‘warning on’ versus ‘warning off’, and pick the order that spares the behavior you care about most.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →If you now give a single warning before a loud noise or mild correction, try splitting the warning: two-second tone, brief event, two-second tone again, and watch if the client’s baseline responding stays smoother.
02At a glance
03Original abstract
Long-Evans rats were exposed to a succession of conditioned-suppression procedures involving pairings of (1) signal-shock, (2) shock-signal, and (3) a signal-shock-signal sequence in which first and second signals were at first physically identical. Traditional suppression of food-reinforced responding was obtained under the signal-shock arrangement, and exposure to the shock-signal sequence resulted in conditioned enhancement of responding during the signal. The signal-shock-signal condition reliably suppressed responding during the first signal, but produced no differential effect on response rate during the second signal. Baseline responding was least changed from preshock rates under the signal-shock-signal procedure, but baseline rate was considerably reduced under the signal-shock and shock-signal arrangements, the latter yielding most substantial reductions. A second experiment indicated that the magnitude and direction of changes in baseline responding reported in Experiment I were not confined to cases in which the first and second signals in the signal-shock-signal arrangement were physically identical. It is suggested that the major effects of the conditioned-suppression procedure on response rate might not be confined to presentations of the signal.
Journal of the experimental analysis of behavior, 1976 · doi:10.1901/jeab.1976.25-263