Comparing two variations of the high‐probability instructional sequence to improve food consumption with a child with autism
Either style of high-probability sequence works for food refusal—just let the parent pick.
01Research in Context
What this study did
Trejo et al. (2018) worked with one child with autism who ate almost nothing at meals.
They tested two kinds of high-probability (Hi-P) sequences. Both start with easy requests the child always obeys.
The ‘similar’ version used eating-related requests first. The ‘dissimilar’ version used play requests first.
An alternating-treatments design showed which style put more food in the child’s mouth.
What they found
Both Hi-P styles tripled food intake and cut screaming, spitting, and head-turning to near zero.
Parents then picked the style they liked and faded the sequence out. Gains stuck for weeks.
Topography did not matter; parent choice did.
How this fits with other research
Older DTT papers already said variety keeps kids with autism engaged. Wing (1981) rotated reinforcers within lessons. Allison et al. (1980) switched tasks every few trials. Both saw better accuracy and happier kids.
Trejo et al. (2018) extends this idea to feeding. They show that variety inside the Hi-P sequence itself is not required; either route works as long as momentum starts.
Wilson et al. (2020) and Bloh et al. (2023) used the same alternating-treatments logic to compare video-modeling styles. Their mixed results echo Trejo’s main point: test both, then let individual data or parent preference pick the winner.
Why it matters
You no longer need to hunt for the ‘perfect’ Hi-P request type. Run a quick comparison of similar versus dissimilar easy tasks, graph the bites, and let the family choose. Start Monday with five easy requests the child already loves, follow with one bite prompt, and you have an evidence-based feeding jump-start that fits real-life therapy time.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Alternate two Hi-P warm-ups at lunch, track bites, and ask the parent which feels easier.
02At a glance
03Original abstract
The present study evaluates the effects of two variations of the high‐probability (Hi P) instructional sequence with a child with autism spectrum disorder. In one variation, the Hi P task was topographically similar to the low‐probability task of food consumption, whereas the second variation involved Hi P tasks that were not topographically similar to the low‐probability task. Results show that both variations of the Hi P sequence were effective at improving consumption and reducing inappropriate mealtime behavior. The participant's mother chose a preferred variation of the Hi P intervention to pursue systematic fading with, the topographically similar Hi P sequence. Consumption remained high and inappropriate mealtime behavior remained low throughout the fading process. Implications for continued research and practice in the area of feeding intervention are provided.
Behavioral Interventions, 2018 · doi:10.1002/bin.1639