The effects of warned loss magnitude and timeout duration on human avoidance behavior
Bigger warned losses boost avoidance, longer timeouts shrink it, and meaningless warnings collapse the behavior.
01Research in Context
What this study did
Ono and colleagues ran a lab game with college students.
Each person pressed a lever to win points and avoid losing them.
The team changed three things: how big the warned loss was, how long the timeout lasted, and whether the warning signal really predicted the loss.
What they found
Bigger warned losses made people avoid more, not less.
Longer timeouts did the opposite—people eased off when the break dragged on.
When the warning light meant nothing, avoidance almost stopped.
How this fits with other research
WEINER (1964) already showed that larger point losses turn escape into avoidance; Ono confirms the rule and adds that timeout length matters too.
Baron et al. (1968) saw a warning stimulus first help, then hurt avoidance; Ono pins the drop on low predictive value, showing the signal must mean something.
Richardson et al. (2008) found that rats worked harder to avoid short timeouts; Ono repeats the pattern with humans, linking rat data to our clinic world.
Toegel et al. (2022) hint that rich reinforcement makes timeout feel worse; Ono’s longer-timeout drop fits—when the break feels endless, people stop trying.
Why it matters
You set response-cost or timeout programs every day.
This lab map says: make the loss clear and sizable if you want fast avoidance, keep the timeout brief so the client stays motivated, and always pair your warning cue with the real consequence.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Cut your timeout to 30–60 s, raise the point loss to a level the client notices, and be sure your warning word or card always precedes the loss.
02At a glance
03Original abstract
This study experimentally investigated the determinants of avoidance behavior when participants are forewarned of aversive outcomes. The effects of 3 variables on avoidance behavior were examined: point-loss amount (5 levels, from 20 to 100 points), duration of timeout from positive reinforcement (5 levels, 20 to 100 s), and 3 predictive accuracy levels (100%, 50%, and 0%) of warning stimuli. Twelve participants completed 3 sessions, each comprising 25 discrete trials, that differed in predictive accuracy level. Throughout a session, a participant engaged in button press responses that were reinforced by points under a conjunctive fixed-ratio fixed-interval schedule. During each trial, a warning stimulus that indicated a loss amount and a timeout duration was presented. If the participant pressed the avoidance button, then the timeout started, otherwise the loss occurred. The trial ended with termination of timeout or an occurrence of the loss. Results showed that avoidance responses increased when the loss amount increased and decreased when the timeout duration increased. The frequency of avoidance responses was lowest when the predictive accuracy of warning stimuli was 0%. These findings demonstrated that this experimental procedure could be useful for investigating human avoidance behavior outside the laboratory.
Journal of the Experimental Analysis of Behavior, 2021 · doi:10.1002/jeab.658