Successive interresponse times in fixed-ratio and second-order fixed-ratio performance.
Guaranteed food creates a dip-peak in response speed right before delivery; uncertain food smooths it out.
01Research in Context
What this study did
Scientists watched how fast rats pressed a lever under two setups. One group worked on a simple fixed-ratio 25 schedule — every 25 presses earned food. The other group worked on a second-order schedule — rats had to finish five small FR-5 blocks, but food came only after the last block.
What they found
On simple FR, the time between presses dipped low, then spiked high, right before food arrived. On the second-order schedule, that dip-peak pattern vanished. Latency to the first press also shrank as food got closer under simple FR.
How this fits with other research
Thomas (1968) first showed that response rate under fixed schedules is built from orderly mini-patterns. The 1969 paper sharpens that idea by proving the dip-peak is tied to certain food, not just ratio size.
Rogers-Warren et al. (1976) later added a visible clock to human FI tasks and saw smoother, slower responding. Their tweak shows external cues can reshape the same within-interval wiggles Davison (1969) found in rats.
Catania et al. (1972) tracked accuracy, not speed, within FI schedules and saw a mid-interval dip. Together, these studies say: whatever you measure — speed, accuracy, or latency — behavior breathes inside the schedule.
Why it matters
When you see a client rush then pause before the reinforcer, check if that reinforcer is guaranteed or only occasional. If praise or tokens come every time, expect the dip-peak; if they come only after many chunks, the pattern may flatten. Use this to decide when to thin or chunk schedules without losing momentum.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Count interresponse times during the last five responses before reinforcement; if you see a dip then spike, try breaking the ratio into smaller, intermittently reinforced chunks.
02At a glance
03Original abstract
Three rats were trained on a schedule in which every sixth response produced a timeout of 5 sec minimum duration, and food was delivered at the onset of timeout. Successive interresponse times were measured under these conditions, and also when behavior was maintained by second-order fixed-ratio and fixed-interval schedules. Under the second-order schedules, each six-response fixed-ratio component was followed by a timeout, and occasionally food was delivered at the onset of a timeout. In the fixed-ratio schedule, the successive interresponse times showed a decrease followed by an increase before food delivery, but this systematic variation in interresponse times was not found when the performance was under second-order reinforcement. Under both second-order schedules the latencies of successive components, and the successive interresponse times within each component, showed a decrease as food delivery was approached.
Journal of the experimental analysis of behavior, 1969 · doi:10.1901/jeab.1969.12-385