FREQUENCY DISTRIBUTION OF INTERRESPONSE TIMES DURING VI AND VR REINFORCEMENT.
After the first response, VI and VR schedules feel the same to the learner, so you can choose either for practical reasons.
01Research in Context
What this study did
KINTSCH (1965) watched how fast responses came after each reinforcer on VI and VR schedules.
The team measured the tiny gaps between responses, called inter-response times (IRTs).
They wanted to see if timing patterns differ once the first post-reinforcer response is out of the way.
What they found
After the first response, the IRT pictures looked almost the same under VI and VR.
There was no clear chain effect; one short gap did not predict the next.
In plain words, the two famous schedules feel the same to the organism once responding is rolling.
How this fits with other research
Delamater et al. (1986) later showed adult humans also treat VR like VI plus a feedback loop, giving the same response rate.
Lincoln et al. (1988) then wrapped both schedules into one math model, turning the 1965 observation into equations you can plug into a spreadsheet.
Davison et al. (1984) warn that concurrent VI VR "matching" can be an artifact of the schedule itself, so equal IRTs do not always mean equal choice rules.
Why it matters
If VI and VR produce the same micro-timing, you can pick either schedule for ease of use without fear of hidden response drag.
Swap in VR when you need faster acquisition, stay with VI when you want steady maintenance, and expect similar post-reinforcer pauses either way.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Track the first three response latencies after reinforcement; if they look alike across your VI and VR targets, keep the schedule that is easier to manage.
02At a glance
03Original abstract
A detailed analysis was made of the interresponse times (IRTs) of two rats under both a VI 40-sec and a VR 15-sec schedule. Except for the latency of the first response after a reinforcement, the mean IRTs of all further responses differed little. Similarly, the frequency distributions of the successive IRTs did not vary greatly, but were of no simple form. Sequential dependencies between successive IRTs were small, never accounting for more than 1% of the variance.
Journal of the experimental analysis of behavior, 1965 · doi:10.1901/jeab.1965.8-347