Sidman avoidance in the fish.
Sidman avoidance produces predictable, negatively accelerated responding in goldfish, and warning cues only redistribute, not remove, the crossings.
01Research in Context
What this study did
THOMAS et al. (1963) put goldfish in a tank with a shuttle barrier.
If the fish did not cross within 20 s it got a brief shock.
Crossing started a new 20-s clock.
The team added a light or tone before some shocks to see how warnings changed crossings.
What they found
The fish quickly learned to cross.
Their rate rose fast early in the session and then leveled off.
Warning signals did not stop crossings.
Instead, fish waited until the cue came and then crossed, bunching responses near the end of each interval.
How this fits with other research
BOLLEHOFFMAN et al. (1964) showed that animals need immediate feedback, not just shock delay, to learn Sidman avoidance.
The fish study adds the first clear baseline: once learned, the pattern holds even in a simple nervous system.
Thomas (1968) later removed shock from rats and saw crossings stop in under four hours.
Together the three papers trace the full life span of Sidman behavior: acquisition, steady state, extinction.
GRAF et al. (1963) ran a similar setup with pigeons and saw the same cue-induced redistribution, suggesting the effect crosses species.
Why it matters
You now know that Sidman schedules create reliable, negatively accelerated response curves even in animals with tiny brains.
If you use free-operant avoidance with clients, expect early bursts of responding that settle into a steady rhythm.
Adding a warning stimulus will shift, not eliminate, the behavior.
Use this to time prompts or to thin schedules without losing the avoidance entirely.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Plot your client’s crossings minute-by-minute; if you see the fish-shaped curve, keep the schedule—if not, check for missing response feedback.
02At a glance
03Original abstract
Goldfish trained in a shuttle box under conditions in which changing compartments postponed shock for 20 sec showed a substantial, negatively accelerated increase in rate of crossing. That the avoidance-contingency was responsible for the change in behavior is suggested by the fact that no significant increase in rate of crossing appeared in control animals which were paired with the experimentals and shocked whenever the experimental animals were shocked; there is some evidence, indeed, that the control animals were handicapped in their subsequent adjustment to the avoidance condition. The introduction of a warning stimulus (light) in the last 5 sec of the response-shock interval decreased the rate of crossing in the first 15 sec and increased the rate of crossing in the last 5 sec. Reducing the shock-shock interval from 20 sec to 2.5 sec had no marked effect. The results are compared with those obtained in analogous experiments with higher animals.
Journal of the experimental analysis of behavior, 1963 · doi:10.1901/jeab.1963.6-47