Nd Project Example 1 [623386]

BEHAVIOURAL ASSESSMENT AND TREATMENT OF
FOOD REFUSAL IN CHILDREN WITH AUTISM

1

ABSTRACT
Escape extinction combined with differential reinfo rcement of alternative behaviour has been
proved to be an effective treatment for food refusa l when implemented in the food selectivity of
children with developmental disabilities . In the current study we tried to determine whether the
texture of food increases the possibility of food r efusal in children with autism. The study had the
objective to decrease food refusal by using differe ntial reinforcement of alternative behaviour and
escape extinction. First, a behavioural assessment was conducted in order to determine the
relationship between food refusal and food texture. Further, we implemented the first condition of
the treatment that consisted of increasing toleranc e for the food presented by breaking the eating
task into 8 steps. Escape extinction and differenti al reinforcement of alternative behavior were
implemented. In the second condition the same proce dures were implemented for mouth clean.
Differential reinforcement was delivered on a VI sc hedule to increase food acceptance for the food
types presented in the first condition. Results of the intervention showed a decrease in food refusal
and a significant increase in bite acceptance.
DESCRIPTORS: functional assessment, escape extincti on, differential reinforcement of alternative
behaviour, food refusal, food acceptance.

2

Food refusal occurs frequently among children with developmental disabilities (Riordan, Iwata,
Finney, Wohl, & Stanley, 1984) and children with au tism. Feeding problems of this nature can have
as a result health problems, such as developmental delays and malnutrition. Studies show that 80%
of individuals with severe or profound mental retar dation exhibit these problems (Perske, Clifton,
McClean and Stein, 1977).
Even if the cause varies or is unknown, feeding pro blems may persist as a result of environmental
factors. Specifically, permitting escape contingent on resistance and other problem behaviors may
increase and maintain food refusal and selective ea ting. The literature has demonstrated the
effectiveness of a variety of behavioral interventi ons for increasing food acceptance and
consumption. Intervention components have included escape-extinction, differential reinforcement
of alternative behavior, differential reinforcement of incompatible behavior, choice, and contingent
reinforcement in the form of attention, preferred f oods, or preferred materials (Cooper et al., 1995;
Riordan et al., 1984).
Many studies had as objective treating feeding prob lems in individuals with autism or other mental
disabilities: Sevin, Gulotta & Sierp in 2002 examin ed the effects of sequentially introduced
treatment across multiple topographies of food refu sal. Treatment with non-removal of the spoon
produced an increase in food acceptance and a decre ase in disruption, but expulsion of food
increased.
Riordan et al. (1984) conducted a behavioural asses sment on chronic food refusal in
developmentally disabled children. They examined th e eating behaviour of four children with no
self-feeding skills. Treatments consisted of: socia l praise, access to preferred foods, brief periods of
toy play and forced feeding. Results of multiple-ba seline and reversal designs showed behavioural
improvements for all the subjects and increased foo d acceptance.

3
Patel et al. (2002) evaluated the effectiveness of two differential reinforcement procedures on food
acceptance versus mouth clean. The participants at the study were three children who have been
diagnosed with a feeding disorder. With the use of differential reinforcement only, the food
acceptance level remained unchanged. Acceptance lev el and mouth clean level increased only when
differential reinforcement and escape extinction pr ocedure were applied. After removing the escape
extinction procedure, the acceptance was maintained for 2 out of 3 participants.
Piazza et. al. (2003) compared the effects of posit ive reinforcement alone, escape extinction alone
and positive reinforcement with escape extinction i n the treatment of the food and fluid refusal of 4
children who had been diagnosed with paediatric fee ding disorder. Food refusal did not decrease
when the procedures mentioned above when implemente d independently. By combining the two
procedures they succeeded in increasing consumption of fluid and food for some participants and
decreased inappropriate behaviour.
Differential reinforcement only may not be enough f or children who display total refusal. Hoch et al.
(1994) considered that reinforcement for acceptance alone was not effective in increasing food
acceptance. A significant increase in acceptance wa s observed only when appropriate behaviours
were reinforced and the inappropriate behaviours no longer produced escape. The negative outcome
for this intervention was that food expulsion incre ased. Therefore Hoch et. al. changed the
contingencies into DRA for swallowing.
The purpose of the present study is to decrease foo d refusal in children with autism by using
differential reinforcement of alternative behaviour and escape extinction. First, a behavioural
assessment was conducted in order to determine the relationship between food refusal and food
texture. Further, the types of food for which food refusal occurred was selected and included in the
subsequent treatment procedure. First, we examined the extent to which differential reinforcement

4
of alternative behaviour (DRA) for food acceptance and escape extinction resulted in increases in
food or liquid consumption, as the procedures were applied on every step of the eating task
individually. Second, we selected the food items fo r which all the 8 steps were being completed
individually by the subject, and put them under the second condition: DRA for mouth clean and
escape extinction.
METHOD
Participants. The participant of this study was a 7 year old chil d who had a diagnosis of autism. He
manifested partial food refusal. At the time the st udy was made, he attended a centre where he was
benefiting from behavioural intervention based on t he principles of ABA, 5 hours per day, 5 days
per week and was a student at a mainstream kinderga rten where he attended 2 hours per day. The
child was presenting self-feeding skills and all th e appropriate behaviour for accepting food: sitting
at the table, holding fork, taking food from the pl ate, grabbing food, bringing food close to mouth,
smelling food, putting food in mouth, chewing food, ingesting food. He was accepting food types
such as: cheese, yoghurt, pizza, cocoa milk, French fries, sweets, bananas etc. Non-accepted food:
soup, sauce, meat, rice, bacon, mushrooms, olives, milk, apple, pear, melon, potato etc. The food
refusal was associated with behaviours such as expu lsion of food, pushing the plate or the fork
away, turning away from the table, leaving the tabl e, screaming, crying. Some of these behaviours
appeared either when he was entering the kitchen or when the food was presented.
Setting. The feeding sessions were conducted in the kitchen of the center where his ABA therapy
sessions took place. The therapist and the subject were present in the kitchen during the session.
Two to three sessions took place every day, 5 days per week.

5

Materials. Data were collected every session using paper and p encil. The type of spoon used for the
child had the regular size and the quantity of food presented was variable, depending on the type of
food.
Behavioural definitions. Food refusal was defined as any instance of: (a) Ch ild turning his head past
his shoulder when the food was presented, pushing f ork and/ or plate away, leaving the table when
the food was presented; (b) Expulsion (any amount o f food that had been in the mouth and became
visible on the lip and/ or chin area); (c) Sobbing, screaming, for at least 5 seconds during the
interval.
Data collection. The data was collected on frequency of challenging behaviour. We measured the
number of occurrences of food refusal per session. The session ended after 5 minutes or after the
subject finished all the food presented. Data on fo od refusal occurrence was converted to percentage
by dividing the total number of refusal occurrences by the number of food presentations in one
interval, multiplied by 100.
Interobserver agreement. A second observer collected data during 45% of the sessions during
behavioural assessment and 28% of the treatment ses sions. An agreement was scored if both
observers recorded any of the behaviours above occu rring at any time during the interval.
Agreement was calculated by dividing the number of agreements by the number of agreements plus
disagreements and multiplying by 100%. IOA for the number of presentations was 100% for all
sessions. IOA for food refusal was calculated for e ach condition as follows: for functional
assessment mean agreement was 92% (range 82% to 100 %); for solid food agreement was 84%
(range 82% to 86%), for liquid was 96% (range 93% to 98%), for mashed the agreement was 90%
(range 81% to 97%) and 96% (range, 91% to 100%) for mixed texture. For treatment sessions mean

6
agreement was 92% (range 83% to 99%). For solid , agreement was 95% (range 91% to 100%), for
liquid it was 94% (range 82 to 100%), for mashed it was 89% (range 82% to 100%) and for mixed
food the agreement was 89%.
Experimental design. A reversal design combining differential reinforcem ent of alternative
behaviour for acceptance (DRA-F.Ac.) and escape ex tinction (E.E.) versus differential
reinforcement of alternative behaviour for mouth cl ean (DRA-M.C.) and escape extinction was used
to decrease food refusal and increase bite acceptan ce. Therefore, solid food was assessed on a
ABCBCBC design and the rest of the food types were assessed on a ABCBC reversal design.
The procedure used to decrease food refusal is DRA (Differential Reinforcement of Alternative
Behaviour) and escape extinction procedure. DRA is applied by reinforcing a behaviour that
provides a desirable alternative to the problem beh aviour, but is not necessarily incompatible with it .
In the present study we used DRA to reinforce each new step of food acceptance individually, on
continuous or an intermittent reinforcement schedul e. Prior to each session, participant was given a
preference assessment to determine the most effecti ve reinforcers. When food refusal occurred, an
escape extinction procedure was implemented.
Procedure
Behavioural assessment. In order to determine the relationship between food refusal and food
texture, we manipulated the antecedent conditions ( food textures). The behavioural assessment was
conducted for 5 days, 4 sessions per day, in which we assessed the refusal of different food types,
divided into 4 categories of food textures: liquid, mashed, solid and mixed.
For each food texture we chose 5 nourishments: liqu id (chicken soup, compote, milk etc.), mashed
(mashed vegetables, whipped cream, mashed fruits et c.), solid (rice, meat, mushrooms etc.), mixed
(pizza, vegetables salad, etc.). Each food was asse ssed for 5 minutes. One trial represented holding a

7
tea spoon of the food type in front of participant’ s lips for 5 seconds . After the 5 seconds interval
ended, there was a 5 to 10 seconds break, which var ied depending on the participant accepting the
food.

Figure 1. Percentage of trials with food refusal fo r the behavioural assessment
Results
The results show that all the food textures present ed constituted a problem of acceptance for the
subject. For solid textures, the subject displayed food refusal for 75% of the sessions, for mashed
and liquid food the food refusal was 100%, and for mixed food, the refuse appeared in 77% of the
trials. The foods selected for food refusal were ty pes of food that the parents claimed he never or
only sometimes accepted at home when they were pres ented to him. The parents also mentioned that
the behaviours correlated to food refusal produced escape in 100% of the situations.

8

Treatment
During Condition 1 (DRA-F.Ac. + E.E.), we divided t he food acceptance behaviour into the
following steps: touching food, picking up food, br inging food close to mouth, smelling food,
licking food, Biting food, putting food to mouth, c hewing food, ingesting food. If the first response
in the hierarchy above appeared in 5 seconds, the i nterest in food has been proved, so the instruction
associated with the following step was given by the instructor. The starting target for each food was
established depending on the first successful respo nse within the response hierarchy.
The treatment sessions were conducted daily, once a day, 5 days per week, before the one to one
teaching sessions. One session consisted of the ins tructor modelling the desired response and give
the instruction “Do this!” the child had 10 second s to imitate the response. If needed, physical
guidance was used. If the acceptance behaviour occu rred, the reinforcer was delivered for 20
seconds. All the other steps that the child could p erform individually were reinforced on a FR-2
schedule of reinforcement. , by beginning alternate ly with the 1 st or the 2 nd step. In case of refusal,
the inappropriate behaviours did not produce escape . That is, the instructor continued with the
request of the participant responding correctly, an d prompted him physically and/ or verbally on a FI
-10s. Sessions ended after 10 minutes or when the c hild finished the last step. The step was
considered mastered when the child had 5 independen t responses on the target behaviour. One food
type was presented in each session. For each presen tation we followed every step from the hierarchy
and collected data on food refusal. In the first se ssion we assessed the solid food, in the second
session we assessed the liquid food, in the third o ne the mashed food, and in the last session the
mixed food and then repeated them until the food re fusal decreased to 0.

9
In the second condition (DRA-M.C. + E.E) the food a ccepted following Condition 1 was presented,
but reinforcement was only delivered for mouth clea n.
Feeding sessions (DRA-F.Ac + E.E.) were conducted b y the ABA supervisor (the investigator).
Maintenance sessions (DRA-M.C. +E.E.) were conducte d by children’s own individual ABA
therapists.
The number of food types presented for each texture were different because of the short time we had
until the end of the study and because of the unava ilability of the subject for 3 weeks during the
study. Therefore, for solid food we applied the tre atment for 3 different foods, and for the other 3
textures (liquid, mashed, mixed) only 2 types of fo od.
Reinforcers were determined every day, prior to th e treatment session, by conducting a preference
assessment.

Figure 2. Percentage of trials with food refusal fo r solid texture

10

Figure 3. Percentage of trials with food refusal fo r liquid texture

Figure 4. Percentage of trials with food refusal fo r mashed texture

11

Figure 5. Percentage of trials with food refusal fo r mixed texture

12

RESULTS
During the first condition, the differential reinfo rcement of food acceptance produced a rapid and
significant decrease in food refusal. The data disp layed in figure 2 show the percentage of food
refusal per trial for solid texture.
During the behavioural assessment, the refuse for solid food was 75% (Figure 1). During condition
1, the food refusal decreased to 43% in the first s ession and continued to decrease until it reached
0% in the subsequent sessions. For food 2, the refu sal was 8% in the first session and then decreased
to 0%. Food 3 had a range of refusal from 75% to 0% . For all 3 foods. The food refusal percentage
remained to 0% level in the second condition. The liquid texture results (Figure 3), also show a
significant decrease of refusal, from 100% during t he assessment to 40% in the first treatment
session. Refusal remained low for the next sessions , for both foods and during both conditions.
Mashed texture, displayed in figure 4 proved to cause 100 % refusal during the behavioural
assessment and had a significant level of acceptanc e during both conditions (refusal range: 11% to
0%). The data for mixed textures are displayed in figure 5. A level of 77% refusal was displayed
during the assessment, which decreased to 0% in the first condition and remained low in the
maintenance condition. For Food 2 we only implement ed the second condition, and the level of
refusal remained at 0%.
DISCUSSIONS
In the current investigation we compared the food r efusal level for 4 textures: solid, liquid, mashed
and mixed. While food refusal had a high level befo re intervention, as data from the behavioural
assessment proved, the level of food acceptance inc reased significantly and the level of refusal
decreased to 0% during treatment. In one condition, the eating task was divided into 8 steps of

13
acceptance for each target. In the second condition , the maintenance conditions, the foods that
reached the 100% level of acceptance were presented , and reinforcement was only delivered for
mouth clean.
There are few possible explanations for the results of this study. Firstly, the 8 steps we established
for the eating task, made the task easier for the c hild, as the reinforcement was delivered for easier
tasks, such as touching food or putting food close to the mouth, rather than at the end of the eating
process. Also, differential reinforcement of differ ent levels of acceptance helped the child learn the
task easier and expose to new textures and food typ es that he did not accept before.
Secondly, the escape extinction procedure may have played an important role in the treatment of
food refusal. Prior to the treatment, the behaviour s correlated to food refusal produced escape from
the food presented, by changing it with another typ e of food or by being allowed to leave the table.
During treatment, the food refusal behaviours were put under extinction, and the intervention
continued until the criteria were met. These findin gs are consistent with the ones described in Patel
et. al.’s study from 2002 who evaluated two differe ntial reinforcement procedures with escape
extinction. Patel et. al. considered that bringing the behaviour of acceptance into contact with the
reinforcement contingency, without giving the child the opportunity to escape from the task was
responsible for food acceptance behaviour.
One limitation of this study is that the procedure described above might function only in children
with certain eating skills and with partial food re fusal. For children with no feeding skills, the
modelling procedure cannot be as efficient without teaching them these skills beforehand. Also, the
small number of food presented in this study does n ot give us the certainty that the child will extend
the food acceptance on any kind of food or texture. For future replications, the aspect of a larger
variety of food should be taken into consideration.

14
In sum, the data presented suggest that differentia l reinforcement along with the escape extinction
procedure were an effective strategy in decreasing food refusal and increasing food acceptance.
Future studies should also evaluate the efficacy of escape extinction with lower reinforcement rates
or with escape extinction only.

15

REFERENCES
Cooper, L. J., Wacker, D. P., McComas, J., Brown, K ., Peck, S. M., Richman, D., et al. (1995). Use
of component analysis to identify active variables in treatment packages for children with feeding
disorders. Journal of Applied Behavior Analysis , 28, 139–153.
Hoch, T.A., Babbit, R.L., Coe, D.A., Krell, D.M. & Hackbert, L. (1994). Contingency contacting:
Combining positive reinforcement and escape extinct ion procedures to treat persistent food refusal.
Behavior modification, 18, 106-128.
Patel M.R., Piazza C.C., Martinez C.J., Volkert V.M ., Santana C.M. (2002). An evaluation of two
differential reinforcement procedures with escape e xtinction to treat food refusal. Journal of Applied
Behaviour Analysis
Perske R., Clifton A., McLean B., Stein J.I. (1977) . Mealtime for severely and profoundly
handicapped persons (new concepts and attitudes). Baltimore: University Park Press
Piazza C.C., Patel M.R., Gulotta C.S., Sevin B.M., Layer S.A. (2003). On the relative contributions
of positive reinforcement and escape extinction in the treatment of food refusal . Journal of Applied
Behaviour Analysis
Riordan M.M., Iwata B.A., Finney J.W., Wohl M.K., S tanley A.E. (1984). Behavioural assessment
and treatment of chronic food refusal in handicappe d children . Journal of Applied Behaviour
Analysis
Sevin, B.M., Gulotta C.S., Sierp B.J., Rosica L.A., Miller L.J. (2002 ). Analysis of response
covariation among multiple topographies of food ref usal. Journal of Applied Behaviour Analysis