Selective punishment early and late in fixed-ratio schedules of food reinforcement.
Punishment delivered early in a response chain suppresses behavior more than the same consequence given late.
01Research in Context
What this study did
Dodd et al. (1977) worked with pigeons on a fixed-ratio 100 food schedule. The birds had to peck a key 100 times to earn grain.
On some ratios the birds got a brief electric shock. Shocks came either early (after the first 15 pecks) or late (after the last 15 pecks). The team compared how each timing changed the birds' pecking.
What they found
Shocks early in the ratio cut pecking far more than shocks late in the ratio. Early punishment almost stopped the birds; late punishment only slowed them a little.
The result held steady across birds and sessions. Timing, not just the shock itself, drove the size of the behavior drop.
How this fits with other research
Dardano (1970) first showed that punishing early ratio responses makes the longest pauses. W et al. narrow that idea to a clean early-versus-late test and confirm the early spot is most powerful.
Dardano (1972) let pigeons choose where shock would occur. Birds usually avoided early shocks, matching the new finding that early shocks hurt most.
HOLZ et al. (1963) found punishment beat extinction in cutting DRL pecking. W et al. add that even within punishment, placement matters: early beats late.
Why it matters
If you use response cost or other aversive procedures, deliver them as soon as the first instance of the problem chain occurs. Waiting until the child has almost finished the task (or the ratio) weakens the effect. Next time you implement a punishment procedure, catch the behavior at the first step, not the last.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Move any response-cost or brief restraint to the very first step of the problem chain.
02At a glance
03Original abstract
Pigeons key pecked for grain on a fixed-ratio 100 schedule; electric shocks occurred intermittently at the fifteenth or eighty-fifth response in the ratio. In Experiment I, shock was at the fifteenth response for two birds, and at the eighty-fifth response for two others, in every sixth, twelfth, or eighteenth ratio. Rate of responding decreased as frequency of shock increased, and the pattern of responding included an increased initial pause and low rates or pause-run sequences that extended further into the ratio when shock was at the fifteenth response than when it was at the eighty-fifth response. Shock early in the ratio engendered longer initial pauses than shock late in the ratio. In Experiment II, four birds responded on a two-component multiple schedule in which shock occurred at the fifteenth response of the third ratio in the presence of a white keylight and at the eighty-fifth response of the third ratio in the presence of a green keylight. The overall rates of responding decreased as shock intensity increased. All four birds responded differentially to the white and green keylights, but with a pattern that varied between birds. In general, punishment reduced the probability of responses that preceded it, regardless of the ordinal position of those responses. Both studies confirm that the probability of responding is reduced less by aversive stimuli produced late in a fixed-ratio than by aversive stimuli produced early in a fixed-ratio.
Journal of the experimental analysis of behavior, 1977 · doi:10.1901/jeab.1977.27-443