Within-session Changes In Responding During Autoshaping And Automaintenance Procedures.
Sparse reinforcement flattens and delays the daily response peak even when the food is free, so watch session length during assessments.
01Research in Context
What this study did
The team watched pigeons during autoshaping and negative automaintenance.
Food pellets dropped at different speeds.
They tracked how pecking changed minute-by-minute inside each session.
What they found
When food came slowly the birds’ pecking curve flattened.
The daily peak also arrived later in the hour.
The same slow-rate flattening appeared in both autoshaping and negative automaintenance.
How this fits with other research
Silverman et al. (1994) saw the same flattening under low-rate FI, VR, and DRL schedules, so the pattern crosses many procedures.
Lowe et al. (1995) showed pigeons already track within-session VI rate changes; F et al. now prove this tracking holds even when the food does not depend on the peck.
Thompson et al. (1974) first showed rats lever-pressing under negative automaintenance; this study moves the question to within-session dynamics in pigeons.
Buitelaar et al. (1999) later warned that this drift can twist demand curves; F et al. gave an early data set that supports keeping sessions short.
Why it matters
If you run preference or reinforcer assessments, remember that even non-contingent food can shift responding within the same session.
A long session with sparse reinforcement may look like satiation or extinction when it is just a natural rate effect.
Keep sessions brief or counterbalance the order so the slow-rate flattening does not hide your treatment effect.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Cut your next stimulus-pairing or preference assessment to ten-minute blocks and compare responding in the first and last two minutes.
02At a glance
03Original abstract
Four pigeons were exposed to autoshaping procedures in which an 8‐second light on a response key was followed by food. Pecks on the key had no scheduled consequences. Subjects were also exposed to negative automaintenance procedures in which a peck on the illuminated key canceled the following food. The intertrial interval varied from an average of 7 seconds to an average of 232 seconds in different conditions. Rate of responding usually changed within sessions during autoshaping. Responding also changed within sessions for the 1 subject that responded during negative automaintenance. The within‐session patterns of responding were flatter, peaked later, and were more symmetrical around the middle of the sessions at lower rates of food presentation, regardless of whether subjects responded on autoshaping, negative automaintenance, or previously reported variable‐interval schedules. These results imply that similar variables produce within‐session changes in responding during both classical (Pavlovian) and operant conditioning procedures.
Journal of the experimental analysis of behavior, 1996 · doi:10.1901/jeab.1996.66-51