S. is a 5 years-old verbal girl diagnosed with autism. She is receiving behavioral training for about 8 months at a private, nonprofit association for children with autism, for an average of 20 hours/ week. She usually spoke in two- to three- word sentences and also followed complex instructions.


The goal of this study is to increase food acceptance for S. who is a picky eater and does not engage in appropriate mealtime behavior. Her parents reported that she consumed a limited number of healthy foods (soup, omelet, escalope, chicken nuggets) and totally refused to eat fruits. The refusal of S. consisted of turning head when the fruit was presented, turning on the seat, saying „I don’t want to eat”, pushing the plate away. 

One of the reasons for which the parents ask to introduce fruits in her menu is that at kindergarten, everyday at snack time, all the kids received a fruit to eat and S. then engages in behaviors maintained by attention, refuses to eat and wants to go. Another reason is to have a healthier meal. Each of the foods included in the procedure were never approached and consumed. During the study, S. sat in a regular-sized chair, was a self-feeder and did not have difficulty chewing or swallowing foods after she accepted them.


S.’s sessions were conducted in three different rooms of the association, to get used to different settings from the beginning of the intervention and to promote generalization of the settings. During the sessions, in the room were present the participant, her teacher and, sometimes, the person who videotaped the sessions. The experimenter who implemented the procedures was the tutor who was working daily with the child and sessions were conducted everyday, except Saturday’s and Sunday’s, during regular snack time. In a day, the experimenter conducted minimum 1 session, maximum 3 sessions, depending on the other behavior interventions or schedules that had to be in place.

Materials consisted of food brought from home by S.’s mother (banana, orange, pear, biscuits, stick) and those necessary for reinforcement delivery (jelly bear, chocolate, candies, lollipops etc.), data sheets, a pen and a timer.  

Observers collected data on the acceptance of high-p foods (preferred foods) and low-p foods (non-preferred foods). Frequency of acceptance or compliance with low-p foods or high-p foods (i.e., taking the entire bite within 5 s of presentation, with the fork or with her fingers, putting the food in her mouth and swallowing the food without vomiting) was collected. We gradually increased the number of fruits consumed. 

S.’s mother identified bananas, pears and oranges as food she would like her daughter to eat because those are fruits she often received at kindergarten and the family used to eat but S. refused. These fruits were used as low-p foods.

Foods that she consumed consistently and that can be combined with fruits were identified by S.’s mother as biscuits and sticks. These fruits were used as high-p foods.  

In the high probability sequence treatment, the goal was to increase S.`s acceptance of new foods. The experimenter conducted minimum 1 session, maximum 3 sessions per day, 5 days per week, in which he presented three rapid (approximately every 3- 5 s) presentations  of high-p foods (biscuits or sticks), followed by one bite of low-p (the targeted fruit). Each trial consisted of four bites (3 high-p and 1 low-p bite), for a total of 5 trials per session. ]

If S. accepted the high-p foods, the therapist immediately delivered verbal praise and after 3-5 s began the next trial. If the low-p foods were accepted, verbal praise and a preferred edible item was immediately delivered, and, after 10 s, we began a new trial. If she did not accept the bite within 5 s of the instruction, the therapist removed the plate and waited 10 s before presenting the next bite or beginning the next trial. 

Mastery criteria was acceptance of the low-p bites for 80% or higher, for three consecutive sessions. After the mastery criterion has been reached, we reduced by one the number of high-p bites presentation systematically (e.g. after reaching the mastery criteria for 3 high-p bites, the number of high-p was reduced to 2), until the high-p sequence was completely removed, for each fruit. When low-p foods were accepted at 80% or higher for three consecutive sessions in the absence of the high-p sequence, we discontinued the intervention.

Results and discussion

Acceptance of high-p was: with biscuits100% and 90% of sticks presentations (80% on the first session and 100% on the second session). Thus, biscuits and sticks were established as a high-p instruction.

Before the procedure, S. did not accept any of the low-p foods. When the high-p sequence was implemented, acceptance of the low-p bites increased. Mean acceptance of banana increased to 85% (range, 60% to 100%) when the high-p sequence was introduced. When intervention was faded by systematically reducing the number of high-probability bites by one, mean acceptance increased to 100% and was maintained when the intervention was removed. 

For pear, mean acceptance increased immediately after the high-p sequence was introduced to 100%. When the intervention was faded by one high-p (from 3Hp to 2Hp), mean acceptance decreased to 93% (range, 80% to 100%). The intervention was then successfully faded and the high-p sequence was removed completely with acceptance at 100%.

With oranges, acceptance improved systematically when the high-p sequence was introduced, with acceptance of the low-p food averaging 73% (range 20% to 100%). As in the case of the other two fruits, the high-p bites presentations were successfully reduced and the intervention was removed completely with acceptance at 100%.

The main purpose of this study was to increase food acceptance of a five-year-old girl using the high-p sequence and verbal praise plus a preferred edible item as reinforcers. The results showed us that using the high-probability sequence procedure and the proper reinforcers may increase food acceptance in the absence of escape extinction, physical guidance or other invasive procedures. For those children who may not tolerate even being in the area of certain foods, look at foods, vomiting or engage in active refusal, the high-p procedure may be ineffective in the absence of escape extinction. 

One important limitation of this case study worth mentioning is represented by the fact that generalization probes were not conducted. Collecting data about the generalization phase (like testing, for example, the ability of the child to eat in a kindergarten setting or at home) and increasing the pieces of the target-food can also be goals for future replications. In any case, S.`s mother reported 4 days after this study was completed, that her daughter ate banana and pear at home, with her brother.