Utilizing antecedent manipulations and reinforcement in the treatment of food selectivity by texture.
Article Type:
Food habits (Research)
Reinforcement (Psychology) (Research)
Autistic children (Behavior)
Autistic children (Food and nutrition)
Najdowski, Adel C.
Tarbox, Jonathan
Wilke, Arthur E.
Pub Date:
Name: Education & Treatment of Children Publisher: West Virginia University Press, University of West Virginia Audience: Professional Format: Magazine/Journal Subject: Education; Family and marriage; Social sciences Copyright: COPYRIGHT 2012 West Virginia University Press, University of West Virginia ISSN: 0748-8491
Date: Feb, 2012 Source Volume: 35 Source Issue: 1
Event Code: 310 Science & research
Geographic Scope: United States Geographic Code: 1USA United States

Accession Number:
Full Text:

Food selectivity by texture is relatively common in children. Treatments for food selectivity by texture have included components such as stimulus fading, reinforcement, and escape extinction. The purpose of the current study was to attempt to treat food selectivity by texture utilizing antecedent manipulations and reinforcement in the absence of escape extinction. The current study utilized texture fading and simultaneous presentation to treat food selectivity by texture in a young boy with autism. Experimenters successfully increased the boy's food consumption using this treatment package in a clinical setting. Then, the boy's parents were trained to implement procedures both in the clinical and home settings.

Keywords: differential reinforcement, feeding problems, food selectivity, parents, simultaneous presentation, simultaneous reinforcement, texture fading

Food selectivity by texture is relatively common in children. In fact, a study conducted by Field, Garland, and Williams (2003) found that 26% of children who participated presented with food selectivity by texture. This number was even higher at 31% for children with autism. Children presenting with food selectivity by texture tend to consume a small variety of textures and reject most novel textures. For example, parents of these children often report that their children are still eating jarred and pureed baby foods.

Treatments for food selectivity by texture have included components such as stimulus fading, reinforcement, and escape extinction (EE). For example, Luiselli and Gleason (1987) treated almost total food refusal in a 4-year-old girl who only consumed milk and occasionally pureed baby foods by gradually increasing the textures of foods, providing reinforcement contingent on consuming foods presented, and using EE in the form of physically guiding the girl to open her mouth by lightly separating her lips with one hand and inserting a spoon.

In another example, Shore, Babbitt, Williams, Coe, & Snyder (1998) treated various feeding problems in four children by gradually increasing the textures of the foods presented to participants, providing reinforcement contingent on accepting and swallowing foods, and using EE in the form of nonremoval of the spoon.

One common feature of these two studies is that in addition to using antecedent manipulations and reinforcement to treat feeding problems, EE was used. EE has been shown to be an instrumental component of treatments for feeding problems (Piazza, 2008), as oftentimes without EE, feeding problems are not eliminated. However, the use of EE can also be accompanied by unwanted side effects such as extinction bursts (Piazza, Patel, Gulotta, Sevin, & Layer, 2003). These effects could potentially adversely affect caregiver satisfaction with procedures as well procedural integrity during parent-conducted sessions resulting in a possible relapse of child behaviors associated with food selectivity.

Given these potential side effects, it is important that techniques that do not include EE be explored in the treatment of feeding problems of a less severe nature. Several interventions have been reported to successfully treat feeding problems in the absence of EE, such as: differential reinforcement of alternative behavior (DRA; e.g., Brown, Spencer, & Swift, 2002; Riordan, Iwata, Finney, Wohl, & Stanley, 1984), simultaneous presentation (e.g., Ahearn, 2003; Buckley & Newchok, 2005; Piazza et al., 2002), peer-modeling combined with differential reinforcement (Sira & Fryling, in press), stimulus fading (Tiger & Hanley, 2006), and noncontingent reinforcement (Wilder, Normand, & At well, 2005). However, published attempts to treat food selectivity by texture without the use of EE are nonexistent in the feeding literature as far as we are aware.

The purpose of the current study was to attempt to treat food selectivity by texture utilizing antecedent manipulations and reinforcement in the absence of EE. The current study utilized texture fading (i.e., starting with low textures and fading in higher textures over time) and simultaneous presentation to treat food selectivity by texture following a baseline texture assessment conducted to determine what combination texture to introduce during treatment.


Participant and Setting

Kaleb, a three-year-old boy diagnosed with autism, participated. Kaleb was a client of an in-home intensive behavioral intervention program and was receiving approximately 30 hours of 1:1 intervention at the time. The supervisor of his behavioral intervention referred him to the first author of this treatment evaluation for feeding services. He continued to receive his prescribed treatment hours of in-home behavioral intervention at the time of participating in this study. Kaleb had previously received feeding intervention services once per-week (for 50-min each session) for one year from an occupational therapist who spent sessions feeding Kaleb jarred baby foods and having him touch higher textured foods with his fingers.

Upon referral to us for feeding intervention, his diet consisted of pureed foods from all food groups. The only foods he would consume at a regular texture were chips, pretzels, and popcorn. Two to three weekly sessions were conducted either in a 10 ft x 15 ft clinic room located in a treatment facility for children with autism or in Kaleb's home during his regularly scheduled lunch or dinner. Kaleb, two experimenters, and one parent (either mother or father) were present during sessions.

Data Collection and Inter observer Agreement

Whether or not the following major dependent variables occurred was scored for each bite presentation and converted to percentage of bites accompanied by these behaviors: acceptance, mouth clean, and inappropriate mealtime behavior. Acceptance was defined as the entire bolus of food passing the plane of Kaleb's lips within 5 s of the spoon being presented to his lips. Mouth clean was defined as acceptance and consumption within 25 s of the bite being accepted as evidenced by visual inspection. Inappropriate mealtime behavior included vocal protests (saying no, whining, crying, or any other negative vocalizations related to the food), expulsions (food the size of a pea or larger visible beyond the border of the lips after acceptance and prior to the next bite presentation), gagging (retching or choking sounds, or in the absence of sound, hyper extending the neck or tensing the neck muscles), and pushing/throwing utensils or food away (included pushing the feeder's hand away when food was on a utensil held by the feeder).

Interobserver agreement (IOA) was collected for 67% of sessions by a trained second observer who scored each of these behaviors independent from the primary observer. IOA was calculated by dividing the number of agreements by the number of agreements plus disagreements for each behavior scored and multiplying by 100%. Agreement averaged 99.09%, 99.77%, and 96.45% for acceptance, mouth clean, and inappropriate mealtime behavior, respectively.

Experimental Design

The experimental design employed in this study was a reversal with a multi-element design embedded in baseline conditions to assess which of four textures were associated with high percentages of mouth clean and low percentages of inappropriate mealtime behavior.

General Procedures. Four foods (one from each food group) were each presented in a semi-random order (determined by the roll of a die with no food being presented more than twice in a row) five times for a total of 20 bite presentations per session. Bites were presented approximately every 30-45 s by an experimenter or parent. Each bite presentation began with a verbal prompt ("open") which was presented simultaneously with a level spoonful of food being presented to Kaleb's bottom lip. Acceptances that occurred within 5 s of presentation were praised. Checks for mouth clean (i.e., Kaleb's mouth was visually inspected after prompting him to open his mouth by saying, "show me all gone" or "say, 'ahh'") occurred 25 s after acceptances. Mouth cleans were praised and access to a highly preferred video (according to a preference assessment) was delivered for 15 s. If a bite was not accepted within 5 s, it was removed for 25 s and the next bite was presented.

Baseline/Reversal Four textures (pureed, wet ground, ground, and chopped fine) were assessed in random order using a multi-element design.

Texture Fading 1, The beginning texture presented was based on the results of the texture assessment conducted during baseline. Specifically, a combination texture composed of 75% of the highest texture associated with mouth clean at 85%-100% during baseline was mixed with 25% of the next highest texture. This mixture was presented until a success criterion of mouth clean during 80%-100% of bites presented and less than 20% with inappropriate mealtime behavior for three consecutive sessions was met.

Texture Fading 2. Following the reversal, the same criterion used in the first phase of texture-fading for introducing a new combination texture was used. However, if no textures met this criterion, then the highest texture associated with mouth clean between 75%-84% was introduced until successful. If no textures met this criterion, then texture-fading continued from the previously successful combination texture (i.e., from 75% current texture / 25% next texture, to 50% current texture / 50% next texture, to 25% current texture / 75% next texture, to 100% next texture; see Shore et al, 1998 for a full description of texture-fading). Upon each success at a texture, probes of remaining full textures were conducted to guide texture-fading, and the same rules just described applied to results from probes.

Texture Fading + Simultaneous Presentation. To enhance the effects of texture fading, which failed to be effective for one of the targeted foods, later stages of treatment involved adding simultaneous presentation to the texture fading procedures. For texture fading combined with simultaneous presentation sessions, all procedures remained the same as in texture fading except that two foods were combined and presented simultaneously on the spoon. Specifically, 75% of a food regularly consumed 100% of trials was combined with 25% of a food failing to be consumed using texture fading alone (food consumed 0% of trials for 4 out of 5 sessions).

Results and Discussion

Figure 1 depicts the percentage of mouth clean for the first (top panel) and second (bottom panel) set of target foods that were introduced. During baseline for the first set of foods (top panel), Kaleb demonstrated mouth clean for 100% of bites of the pureed texture, thus a combination texture of 75% puree and 25% wet ground was introduced until successful. Upon returning to baseline, Kaleb demonstrated mouth clean for 80% of bites of the wet ground texture meeting the criterion for introducing this texture. After success with this texture was met, a subsequent probe indicated that mouth clean was observed for 90% of bites of the ground texture. Thus, a combination texture composed of 75% of ground and 25% of chopped fine was introduced. The criterion for success was not met, because one of the target foods (chicken) was consumed 0% of presentations for 4 out of 5 sessions. Anecdotal observations indicated that Kaleb often asked to eat pears instead of chicken, so a simultaneous presentation procedure was implemented in combination with texture fading (chicken with pear using the same combination texture of 75% ground and 25% chopped fine that was already in place). A short multi-element design demonstrated that mouth clean was occurring at higher percentages with the texture fading combined with simultaneous presentation procedure than the texture fading procedure alone, and the criterion for success was met in six sessions of texture fading combined with simultaneous presentation. During a probe of chopped fine without simultaneous presentation, mouth clean was observed to occur during 75% of total bites and 60% of bites of chicken. Since this was a moderately high percentage of mouth clean for chicken in the absence of simultaneous presentation, the next texture introduced was chopped fine using simultaneous presentation (rather than a combination texture). The criterion for success was not met in eight sessions, and data indicated that Kaleb was also beginning to reject pear when presented alone. Thus, potato (chopped fine) was used in place of pear in the simultaneous presentation with chicken for one session and it was observed that Kaleb then began to reject some bites of both pear and potato when each was presented alone. In an effort to avoid any further decreases in mouth clean of previously swallowed foods, chicken was dropped from the meals and only three foods were targeted for seven trials each and a total of 21 bites. After the criterion for success was re-established with the three foods, Kaleb's parents began conducting sessions both in the clinic and home settings with success.

During baseline for the second set of foods (bottom panel), mouth clean was observed at 100% for the wet ground texture, thus a combination texture composed of 75% wet ground and 25% ground was introduced until successful. The next set of probes indicated that mouth clean was observed at 100% for the ground texture, thus a combination texture composed of 75% ground and 25% chopped fine was introduced until successful. Mouth clean during the next probe of the chopped fine texture was below 75%, thus we continued fading our texture from the previously successful texture by introducing 50% ground and 50% chopped fine until successful. Mouth clean during the next probe of chopped fine was at 100%, thus the chopped fine texture was introduced until successful. Kaleb's parents then began implementing treatment and results were not maintained, so the therapist was brought back in for a few more sessions before reintroducing the parents and reaching successful levels of mouth clean in the clinic and home.

This study demonstrated that antecedent manipulations and reinforcement-based procedures increased mouth clean in the absence of EE for some foods in a young boy with autism presenting with food selectivity by texture. Additionally, presenting full textures in baseline using a multi-element design might lend itself to the identification of appropriate textures to introduce. However, future research should evaluate what level of mouth clean during probes is necessary to suggest that a subsequently introduced combination texture will be successful. Replication of these procedures with more participants is also warranted.

The results of this study provide further support that interventions without the component of EE may be effective in some cases. Future research should explore under what conditions EE seems to be successfully omitted. One contributing factor that may be explored is participant and family characteristics that might lend themselves to the treatment being successful. In the case of this study, Kaleb was accustomed to receiving intensive 1:1 behavioral intervention, thus he was rather compliant with remaining seated during feeding sessions and familiar with the format of discrete trial teaching (i.e., responding to a discriminative stimulus correctly and receiving a reinforcer). His parents were also accustomed to the format of treatment that was used for feeding intervention as it resembled the behavior analytic procedures that were already being used during Kaleb's in-home intervention.

Acceptability of the treatment procedures by caregivers might also be relevant to whether or not the outcome of treatment will be favorable (i.e., parents that find the procedures socially valid may be more likely to comply with instructions for preparing for feeding sessions and for parent-conducted meals). In this case, the acceptability of the treatment procedures was high. Kaleb's mother completed a social validity measure composed of six questions on a Likert scale with a score of 1 being associated with "strongly disagreeing" with and 5 being associated with "strongly agreeing" with positive statements. The overall score obtained for this questionnaire was 30 out of 30, indicating that Kaleb's mother was highly satisfied with the treatment goals, procedures, and outcome.

Unfortunately, we did not collect treatment fidelity data during this study so we are unsure whether high satisfaction was correlated with high procedural integrity. Additionally, we noted that the results of the treatment with Kaleb were not maintained when Kaleb's parents initially began conducting treatment sessions. Anecdotal observations suggested that Kaleb's parents accurately implemented procedures during treatment sessions, thus we believe the lack of maintenance was not because his parents were incorrectly following the procedures associated with his treatment. However, the absence of treatment fidelity data is a limitation of the current study and is something that should be considered in future research conducted in this area.

Regardless of the limitations of this study, there are some interesting findings which are noteworthy. First, our results suggest that texture involves more than the consistency of foods (e.g., dryness). After we were unsuccessful in getting Kaleb to consume chicken breast, we introduced chicken cold cuts in our second set of foods. Perhaps because this format of chicken is more water-based, we were able to increase the texture to chopped fine. This factor should be evaluated in future research.

Second, Kaleb's chewing pattern anecdotally improved from a munching chewing pattern at the start of treatment to a circular rotary chewing pattern by the end of treatment (per an independent occupational therapist report) without directly teaching chewing. This might be a potential benefit of using texture fading to increase the consumption of higher texture foods rather than presenting high texture foods to begin with and treating with EE (as the child would also need to be taught to chew). Future research could formally evaluate chewing as an outcome of texture fading.


Ahearn, W.H. (2003). Using simultaneous presentation to increase vegetable consumption in a mildly selective child with autism. Journal of Applied Behavior Analysis, 36, 361-365.

Brown, J. F., Spencer, K., & Swift, S. (2002). A parent-training programme for chronic food refusal: A case study. British Journal of Learning Disabilities, 30,118-121.

Buckley, S.D., & Newchok, D.K. (2005). An evaluation of simultaneous presentation and differential reinforcement with response cost to reduce packing, journal of Applied Behavior Analysis, 38, 405-409.

Field, D., Garland, M., & Williams, K. (2003). Correlates of specific childhood feeding problems. Journal of Paediatrics and Child Health, 39, 299-304.

Luiselli, J.K., & Gleason, DJ. (1987). Combining sensory reinforcement and texture-fading procedures to overcome chronic food refusal. Journal of Behavior Therapy and Experimental Psychiatry, 18, 149-155.

Piazza, C. C. (2008). Feeding disorders and behavior: What have we learned? Developmental Disabilities Research Reviews, 14, 174-181.

Piazza, CC, 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 Behavior Analysis, 36, 309-324.

Piazza, CC, Patel, M.R., Santana, CM., Goh, H., Delia, M.D., & Lancaster, B.M. (2002). An evaluation of simultaneous and sequential presentation of preferred and nonpreferred food to treat food selectivity. Journal of Applied Behavior Analysis, 35, 259-270.

Riordan, M. M., Iwata, B. A., Finney, J. W., Wohl, M K., & Stanley, A. E. (1984). Behavioral assessment and treatment of chronic food refusal in handicapped children. Journal of Applied Behavior Analysis, 17, 327-341.

Shore, B.A., Babbitt, R.L., Williams, K.E., Coe, D.A., & Snyder, A. (1998). Use of texture fading in the treatment of food selectivity. Journal of Applied Behavior Analysis, 31, 621-633.

Sira, B.K., & Fryling, M.J. (in press). Using peer modeling and differential reinforcement in the treatment of food selectivity. Education and Treatment of Children.

Tiger, J. H, & Hanley, G. P. (2006). Using reinforcer pairing and fading to increase the milk consumption of a preschool child. Journal of Applied Behavioral Analysis, 39, 399-403.

Wilder, D. A., Normand, M., & Atwell, J. (2005). Noncontingent reinforcement as treatment for food refusal and associated self-injury. Journal of Applied Behavior Analysis, 38, 549-553.

Adel C. Najdowski, Jonathan Tarbox, and Arthur E. Wilke Center for Autism and Related Disorders, Inc.

Correspondence to Adel Najdowski, Research and Development, Center for Autism and Related Disorders, 19019 Ventura Blvd. Ste. 300, Tarzana, CA 91356; e-mail: a.najdowski@centerforaurism.com.
Gale Copyright:
Copyright 2012 Gale, Cengage Learning. All rights reserved.