Accumulated‐ and distributed‐reinforcer arrangements in the treatment of challenging mealtime behavior
Let the child choose between bite-by-bit prizes or a saved-up reward—both cut mealtime problem behavior, but kids don’t all like the same deal.
01Research in Context
What this study did
Chen and team worked with three kids who had feeding disorders. Each child tried three ways of earning reinforcement during lunch.
One way gave a tiny toy or praise right after every bite. Another way saved the prize until the child ate five bites. A third way used tokens that traded for a bigger toy after five bites.
All sessions used non-removal of the spoon. The child could leave the table only after finishing the set number of bites.
What they found
Every schedule cut problem behavior and helped kids accept bites. No schedule failed.
Yet each child liked a different setup. One wanted the toy after every bite. Another liked saving up for the big prize. The third did best with tokens.
The take-home: match the schedule to the child, not to the textbook.
How this fits with other research
Fulton et al. (2020) first tested the same save-up versus bite-by-bite question with kids who hit or ran to escape work. They saw less problem behavior when the break came only after 15 tasks. The feeding study flips that result: here, bite-by-bite worked just as well and was often preferred. The difference is the response type. Chewing and swallowing are quick; academic tasks are not.
Kirkwood et al. (2021) showed that when food refusal is driven by both escape and attention, you must add extinction for each function. Chen’s study kept extinction constant (non-removal of spoon) and only changed when praise or toys arrived. Together, the papers say: first block all escape and attention, then fine-tune when you deliver the reinforcer.
Van Arsdale et al. (2024) reviewed 15 recent feeding studies and found non-contingent reinforcement is common but poorly defined. Chen’s work gives a clear, testable way to make reinforcement contingent and still flexible across kids.
Why it matters
You no longer need to guess between bite-by-bite praise or a big end-meal reward. Run a quick alternating-treatments probe across two lunches. Let the child’s data pick the schedule. You will likely see the same drop in gagging, crying, or head-turning no matter which you choose, but the child will finish faster and happier when the schedule matches his or her preference. Start Monday by offering both options during one meal and graph which one produces the most bites with the least fuss.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Run two short lunch probes: one toy after every bite versus one toy after five bites; count bites and problem behavior, then let the child’s data pick the schedule.
02At a glance
03Original abstract
AbstractTreatment for inappropriate mealtime behavior often includes extinction and differential reinforcement of alternative behavior (DRA) in a distributed arrangement in which delivery of brief reinforcer access immediately follows each appropriate mealtime response. Alternatively, DRA may be arranged using accumulated reinforcement wherein delivery of longer, continuous access to reinforcers follows the consumption of multiple consecutive bites. Research has suggested that individuals prefer and perform better under accumulated arrangements in academic settings; however, no research to date has evaluated the efficacy of accumulated arrangements with children with feeding disorders. We compared preference for and efficacy of distributed and accumulated (with and without tokens) reinforcement with nonremoval of the spoon. All three treatments effectively reduced inappropriate mealtime behavior and increased acceptance for two participants and decreased packing and increased mouth clean for one participant. Two participants preferred distributed reinforcement and one preferred accumulated reinforcement without tokens. Findings differ from some previous research, which demonstrated that participants favored accumulated arrangements in academic contexts.
Behavioral Interventions, 2022 · doi:10.1002/bin.1899