Printer Friendly

Towards a molecular analysis of function (1).

This article addresses potential methodological and terminological limitations of Functional Assessment, and proposes an approach we are developing to address these concerns. The authors propose that current methods of assessment and data analysis may yield a specific behavior pattern due to the effects of reinforcement that could occur during the analog conditions themselves. Shaping of identified target behaviors during analog conditions is one possible problem discussed that may result in masking or obfuscation of the "true function." A methodology being developed by the authors to minimize this and other potential problems that incorporates a means of quickly determining effective treatment is presented and discussed.

Key words: functional assessment, molecular, original function, current function

**********

Functional assessment is a hallmark of current behavior analytic practice, appearing in over 100 articles in the Journal of Applied Behavior Analysis since 1981. Treatments derived from functional assessment methodology assist in the reduction of disruptive classroom behavior (Horner & Dodson, 1993), severe self-injurious behavior (Vollmer & Vorndran, 1998), the ingestion of inedible substances (Mace & Knight. 1986), and reducing stereotypy (Kennedy, Meyer, Knowles, & Shukla, 2000). There has been a great breadth and scope of work done in this area over the past two decades with significant progress in the development of the methodology. Functional Assessment has become such an integral aspect of the field of Applied Behavior Analysis that it is now a requirement by the federal government for use in educational programs nationally for exceptional children under IDEA '97. In this article, we are addressing a specific methodological limitation we have seen which may be resolved in part via the use of a molecular analysis approach that we have been developing.

Terminological and Methodological Issues in Functional Assessment

We will address both the terminological and methodological issues in functional assessment, and then present suggestions to address the concerns raised. First, we shall focus on the terminological issues, and second, methodological issues in functional assessment.

Use of the term "Function"

Heretofore, the term "function" in the research literature (Iwata, Dorsey, Slifer, Bauman, & Richman, 1994, Mace, 1994, O'Neill, Horner, Albin, Storey, & Sprague, 1997, Vollmer & Smith, 1996) which has been used to refer to maintaining consequences, may pose complications with respect to use of the term itself. When referring to function, one is unclear as to the basic concept involved. Are we merely referring to any reinforcer that can be readily identified as a maintaining consequence of a particular response class; or does the term function refer to the initial reinforcer that shaped the behavior to begin with? The reason we raise this concern is that we have observed situations in which the "original " reinforcer appears not to be readily evident, and another more evident reinforcer that is also maintaining the behavior was identified as the function. This very problem, which is the primary topic of discussion in this article, is further complicated by the lack of terms to adequately illustrate the issue because it has not previously been a focus of concern by behavior analysts.

As an example, which we will more fully describe below, a 15 year old student in our school diagnosed with autism, began engaging in several challenging behaviors including rubbing his ears, loud vocalizations, leaving class and wandering about the school grounds. The behavior analyst conducted a functional assessment and initially determined that the function of the behavior was teacher/staff attention due to the persistence of the behavior when these personnel approached him and tried to find out "what was bothering him." As it turns out, attention was a current reinforcer that served to maintain these behaviors, however, the "true" or "original" function was escape from discomfort due to problems related to recent medical treatment for an ear infection. Once this medical problem was addressed, the behaviors subsided and were no longer a problem during school. In this case example, the readily evident reinforcer did serve to maintain behavior i.e. current function, and eliminating it was necessary for effective treatment, but it did not reduce the rate of behavior. Until the "original" function was identified, appropriate and effective treatment was not possible.

[TABLE 1 OMITTED]

[TABLE 2 OMITTED]

Proposed Terms

First, use of terminology needs to be addressed before we can meaningfully proceed to addressing the methodological concerns. We propose utilizing the terms "original function," and "current function." "Original function" refers to those reinforcers responsible for the initial shaping and maintenance of the target behavior. Alternatively, "current function" refers to those readily evident reinforcers that have come to additionally maintain the same behavior as that maintained by the original function. Reinforcers that are identified as part of the "current function" may have acquired their reinforcing properties at some point after the original function initially shaped the behavior. One possibility is that such reinforcers may acquire their status as a "current function" during an analog condition. That is, it is possible that a behavior, such as self-injury may have been initially shaped by escape from task, but attention also served to reinforce it during an analog condition. This is not to say that there are not a number of other ways in which a behavior class can come to be controlled by multiple functions. Referring back to the case study, above, the following table will clarify the role of the definitions in identifying the different types or classes of functions.

The situation illustrated in Table 2, item 2, above, teaching replacement skills for "inappropriate attention seeking" is not likely to alleviate the problem behavior in question, and may in fact result in exacerbating the problem.

Summary of Terminological Issues

Since our goal as behavior analysts is to identify the function of behavior that leads to determination of effective treatment approaches, identifying the original function becomes crucial. If one cannot identify the original function of the behavior, then we may not be able to identify the relevant reinforcers that would be necessary to devise an appropriate treatment intervention. Some ways in which using only current function may hinder effective treatment include (a) possibly teaching the wrong replacement behavior; (b) allowing access to the wrong reinforcer, (c) eliminating access to the reinforcer of "choice" (original function) for that individual, and other errors. Therefore, the terms we propose above with regard to function will provide a foundation upon which we can meaningfully discuss functional assessment methodology.

Methodological Considerations

In light of our discussion to this point, we propose that a key purpose of functional assessment should be to identify and distinguish between original function and current function. A further elaboration of the case example used above may help illustrate and clarify this distinction. The student in our educational program we referred to earlier, began to engage in an increase in challenging behaviors including leaving his classroom, roaming the hallways, banging on objects, loud vocalizations, rubbing his ears, and refusing redirection to his program area and task. A functional assessment was conducted, which included review of prior records and direct observation. The behavior analyst observed that these behaviors persisted when staff approached this student and attempted to redirect him to his classroom, or attempted to find out what the problem was. His teacher and behavior analyst concluded that the function of the behavior was attention seeking. The behavior analyst additionally considered the fact that a week earlier he had an ear infection that resulted in his pacing behavior, leaving his assigned classroom, banging on objects, and rubbing his ears. She was informed that he went to his physician's office and received appropriate treatment including an ointment to help reduce the itching and irritation in his ears. As a result of information confirming appropriate medical treatment, the notion that the function of the behaviors could actually have been an attempt by this student to reduce the discomfort in his ears was ruled out in favor of the attention-seeking hypothesis. However, all attempts to address this behavior via various means (such as teaching appropriate means of requesting attention, asking for time with staff at breaks, asking for help, etc.) were unsuccessful.

After a discussion with the behavior analyst, the function was re-evaluated, and the medical issue was once again considered. A check-up with the nurse was rescheduled, and it was determined that there was a build-up of the ointment in the ears that was potentially responsible for discomfort. Due to the student's lack of verbal communication skills, he was unable to indicate this problem to his teacher (i.e. that the treatment itself was responsible for the current challenging behavior). His mother and physician were contacted, and application of the ointment was discontinued, and the ointment in his ears was cleaned out with a swab. After this, the problem behaviors ceased and the student was again on-task, and performing as usual at school.

The lesson we learn from this case example is that our methodologies may lead us to incorrect conclusions about the "true" or original function of the behavior we are trying to treat, and may result in inappropriate or ineffective treatment. The behavior analyst had to distinguish between current and original functions to determine the most effective treatment method, which in this case was a medical solution. It may have been the case that there was an attention component in that the student's behaviors may have also been serving a communication function to alleviate his discomfort, or as a means of coping with the discomfort, both of which are forms of behavior maintained by negative reinforcement. However, this raises additional methodological issues. In the next section of this article, we will identify and discuss several of these issues, and how they may interfere with determining function.

Reinforcement History as a Relevant Factor

We are proposing that when conducting a functional assessment (including functional analysis methods), we may need to focus our efforts on identifying the consequences that have initially resulted in the shaping and maintenance of the problem or challenging behavior (i.e. original function). As mentioned earlier in the article, the process of conducting analog conditions may itself result in new or added reinforcers maintaining the behavior during an analysis that may or may not be relevant to the original function of the behavior. Hence, we see our task in functional assessment ultimately as developing a means of identifying the reinforcers that were responsible in the past history of the individual in shaping and maintaining those behaviors that are identified for treatment (Tatham & Wanchisen 1998; Wanchisen & Tatham, 1991). To the extent that such a complete analysis is possible, there may continue to be limitations in the existing methodology, but through discussion and research, we suggest that further advancements of this powerful methodology can be realized.

Role of Establishing Operations in Functional Assessment

One must consider the roles of establishing operations in the functional assessment and in conducting analog conditions in particular. The reinforcers provided in any given analog condition may serve to alter the establishing operation (EO, cf. Michael, 1993, Schillinger & Blakely, 1994), which may have been responsible for occasioning the emission of the problem behavior being evaluated. We are proposing an approach that allows for a means of assessing function in which the effects of altering EOs, as well as other factors such as effects of particular reinforcement history, can be evaluated on a moment-to-moment basis. Hence, in some instances we may be looking for reinforcers that may quickly, albeit temporarily, decrease behavior rates. The behavior would likely decrease in frequency, or cease due to temporarily altering (reducing or eliminating) the value of the reinforcer, indicating a change of EO function. This decrease may only be seen in a molecular analysis, which also can indicate effective treatment simultaneously, maximizing the efficiency of the analysis.

Molecular Functional Assessment

We refer to the method we are proposing as a "Molecular Functional Assessment." In this approach, we do not collapse data on a large scale and look at summary statistics such as rate measures or averages. Instead, we assess the momentary (i.e. "molecular") changes in behavior, typically in 30 second time frames, to be able to determine effects of environmental factors and reinforcement history as the behavior changes during the actual assessment. This methodology will permit for the fine-grained analysis of data necessary to distinguish current from original function, assist the behavior analyst in determining effects of prior history of reinforcement, establishing operations, and aid in determining the other factors such as effects of Sds on the behavior. In addition, the method includes an evaluation of effective treatment intervention.

Practical Application of the Methodology: Purpose and Procedures

The Molecular Functional Assessment methodology focuses on determining original function by systematically identifying and eliminating any possible current function(s), while also evaluating effective treatment. We applied the molecular functional assessment methodology to assess the behaviors of a child referred to our program who displayed aggressive behaviors and who also had apparent eating difficulties.

METHOD

Subject

A 10 year old male living in the residential program and attending our day school, served as the subject of the experiment. This child, whom we shall refer to as Jimmy, was admitted to the program from home where his adoptive parents were unable to effectively manage his behaviors and medical needs. Jimmy's behavioral concerns as observed in our program consisted of aggression in the form of hitting others, and frequent requests for food. Jimmy was functioning within the moderate range of mental retardation, and his vocabulary was very limited, and included the ability to say "eat." The family and funder requested that the issue of frequently requesting food be addressed as a target behavior in addition to aggressive behavior. This request was based in part on a practical issue, and a health issue due to reported food allergies for many food categories, and concern about his being overweight.

Procedure

During the first 5 minutes of the analog conditions, the experimenter reinforced food mands and ignored aggression. During the next 5 minutes, the experimenter responded to food mands with a verbal "no," and aggression resulted in food delivery.

[ILLUSTRATION OMITTED]

Condition 3 was a repeat of Condition 2. The next condition was the treatment assessment phase. The treatment intervention that was chosen was based on results of the above conditions, in which we concluded that the function of the behavior was access to food via aggression. Jimmy was prompted by the experimenter to engage a task for 30 seconds if he manded for food, which resulted in delivery of a food reinforcer at the end of the period if the child was on task. Jimmy was offered a choice of activities that he typically preferred. This condition was carried out for 5 minutes. In the final treatment phase, which also was implemented for 5 minutes, Jimmy received food reinforcement if he manded, was on task for 30 seconds, and did not display aggression. There were no prompts to engage in the task, and he was given food reinforcement after 30 seconds of on-task behavior. Engaging in the task was Jimmy's choice, and he was offered known preferred tasks based upon prior observation.

RESULTS

Figure 1, below, shows the data obtained from this functional analysis. Note that the conditions were run in very short time periods, and were also conducted in two sessions on two consecutive days, for a total of 30 minutes for all conditions. Inter-rater agreement for the two independent observers was at 100% for all conditions. Note that in condition 1, aggression did not occur when not reinforced, and food mands rapidly increased during the five-minute phase. In condition 2, we see a rapid reversal of this trend, in which aggression rates rapidly increased when reinforced. But the salient feature of this analog condition is that we see a short-lived extinction burst for food manding, with a rapid deceleration by the midpoint of that condition, and then both aggression and manding rates co-vary. This result was replicated the next day when condition 2 was continued for ten minutes, in which we see evidence for a behavior chain of food manding, which if not met, was followed by aggression, which then resulted in food delivery. Due to the strength of the behavior chain, and its resistance to extinction, we concluded that this behavior chain was reinforced in the home setting with food, and was an "original function." These behaviors were the main reasons for Jimmy being referred to our program for treatment, which was substantiated as a stable behavior pattern by our functional assessment method.

[ILLUSTRATION OMITTED]

In the treatment phase that followed the assessment phases (1-3), the purpose was to reduce rates of aggression as well as food mands, due to concerns from the family with regard to eating too many snacks. Recall that in this condition, Jimmy was prompted to engage in a task for 30 seconds, which resulted in delivery of an edible reinforcer if he did so without aggression. Mands resulted in a prompt by the experimenter to engage in the task. Aggression was ignored, and did not result in food reinforcement. The results of this condition were zero incidents of aggression, and very low rates of food manding. The final condition was a straightforward DRA in which 30 seconds of unprompted on-task behavior was reinforced with food, and both aggression and food mands were ignored. In this condition, we saw a short extinction burst for mands, which dropped out by the middle of this short phase, and one instance of the behavior chain of manding, then aggression, both of which were ignored. The data support the idea that prompting on-task behavior contingent on mands appeared to reinforce, and hence, maintain the food mands, which indicates that the DRA procedure was overall more effective, which might have been verified by extending that condition.

DISCUSSION

In this study, we evaluated a unique methodology for assessing data from a functional analysis to determine function. We utilized our knowledge of histories of reinforcement and conditioning (cf. Tatham & Wanchisen, 1998; Wanchisen & Tatham, 1991) to devise a methodology to determine original function. If we determine via multiple observers that there was not a very rapid deceleration of the rate of behavior following an extinction burst in the extinction condition, then we can further con "current function." We have also proposed that there are two areas of concern 1) methodological issues with regard to knowing prior reinforcement histories, and masking or obfuscation by the reinforcing conditions in an analog condition, and 2) terminological issues that arise with regard to the manner in which to speak of two (or more) types of potential functions utilizing behavior analytic terms that are precise and technical, and that addresses different reinforcement histories. We have proposed some conventions for terminological usage to refer to two different types of function: "original function" and "current function." We have also proposed a modification to the methodology in the form of a functional assessment approach which can quickly identify changes in EO effects on reinforcement and which aids in identifying "original function" by conducting pointed, brief analog conditions, conducting a molecular analysis of the data during the actual analog conditions, and then moving on to evaluating effective treatment interventions. The suggestions we are making are intended to stimulate further discussion and hopefully research, particularly on the problem of identifying the effects of prior histories of reinforcement (Wanchisen & Tatham, 1998). If one can determine a means of assessing the effects of prior histories of reinforcement, one has a better opportunity to identify the original function.

To further clarify our approach, we are proposing that identification of the original function of behavior is crucial to more effective and efficient treatment, and to do so, one must determine or eliminate current function. We further propose that behavior maintained by a current function is weaker than that maintained by the original function that was responsible for the emergence of the behavior to begin with. If we eliminate the original function of a target behavior, current function(s) may continue to maintain it, but the behavior will be weaker and much less stable. The reason for this is that current function serves to support behavior, but eliminating the current function is not sufficient to reduce or eliminate rates of the target behavior. If we can find the original function, which is primarily responsible for maintaining the behavior, it will lead to a more effective and efficient treatment intervention. In application, we would expect that if we were to eliminate or alter the effects of current function(s), we may see either a small decrement in rate or frequency of the behavior, or a transient effect consisting of an extinction burst with stabilization at or near the previous rate or frequency of behavior. However, if we eliminate or alter the effects of the original function, we would be more likely to dramatically reduce or eliminate the rate or frequency of behavior very rapidly.

In our case example discussed earlier, the approach assists in confirming the identification of the original function while also evaluating treatment effectiveness to distinguish between current function, teacher attention, and original function, discomfort created by the ointment in the student's ears. Consequently, we can see that eliminating teacher attention may serve to weaken or reduce rates of behavior to some extent. However, because we have not identified the original function (ointment in the ears resulting in discomfort) we will not be able to finish our task of completely eliminating the target behavior, and we will not be efficient in determining the appropriate replacement behavior or effective treatment intervention.

Benefits of a Molecular Approach to Functional Assessment

The molecular functional assessment approach may have several benefits that address these concerns, since it looks at the behavior relative to function in very short periods of time. These benefits will be listed and discussed below.

1) As suggested by the name we chose, the molecular approach assesses function on a moment-to-moment basis (in this study, we look at the progression of behavior in 30-second segments within a given analog condition). Standard methods tend to look at more molar measures such as rates or averages which can make it difficult to evaluate effects of extinction or determine whether a shaping process is occurring. Such analyses are crucial to distinguishing current vs. original function.

2) Each of the analog conditions are alternated with strategic environmental manipulations to determine the original function. The molecular approach will minimize the problem of establishing a new (current) function, and help assess the existence of a current function. By evaluating data on a moment-to-moment basis, the behavior analyst is better able to distinguish patterns suggesting that the behavior being evaluated is resistant to extinction or extinguishes rapidly. When a reinforcer is presented contingently in the initial analog condition that is presumed to be function of the behavior, and in subsequently is withdrawn, a molecular analysis of the data will reveal one of two primary patterns. First, resistance to extinction with an extinction burst would suggest that the behavior was more likely to have been in the person's existing behavioral repertoire (original function). However, if the behavior extinguishes rapidly in this subsequent condition, the conclusion may be made that behavior that was maintained by the reinforcer in the prior analog condition was weak, was more recently shaped, and firm that this behavior was in the existing repertoire of the person (i.e. the result of prior conditioning and maintenance over some time period). Furthermore, we also have a firm basis for arguing that our current analog condition did not result in establishing a new (i.e. current) function. After this demonstration, we also immediately implemented a treatment intervention designed to teach a functionally equivalent replacement behavior, and evaluate its efficacy.

Our conceptualization is that utilizing prior reinforcement history, knowledge about rates of behavior, and response strength, along with a moment-to-moment analysis of behavior, allows for a means to quickly determine function. The argument is that we can determine original function via rapidly changing conditions utilizing the alternating treatments (or multi-element baseline) design (Barlow & Hayes, 1979, Hersen & Barlow, 1978, Ulman & Sulzer-Azaroff; 1975). One indicator of original function is the absence of an acquisition curve. That is, the behavior in question rapidly emerges, and additionally, we see an extinction burst during the extinction session. The empirical data presented here are a demonstration of the utility of our molecular approach to functional assessment, and further demonstrates that it is possible to determine original function directly. In this case, the original function appeared to be the only function in place. Our approach offers a means of minimizing establishment of a current function during the analog conditions; identifying current functions if they exist; and then distinguishing between original and current function as warranted. We are proposing that we have a methodology that can address the issue of uncovering original function when other possible maintaining reinforcers are masking that original function. Magee and Ellis (2000) used a similar approach to eliminate behaviors that occurred in a response hierarchy in which different behaviors were maintained by the same reinforcer. Our approach is similar to that used by Magee and Ellis (2000), however, in our methodology, we identified different reinforcers that were maintaining the same behavior. Due to the promising results we obtained with regard to rapidly determining original function, we see the potential for further research that would replicate and expand upon this method. We have demonstrated the utility of our approach in a relatively straightforward analysis that can be applied to more complex situations such as those discussed in the case example earlier in the article.

Toward a Molecular Approach to Functional Assessment

We have discussed potential difficulties in identifying what we are referring to as "original function" vs. "current function." We have also proposed that there are two areas of concern 1) methodological issues with regard to knowing prior reinforcement histories, and masking or obfuscation by the reinforcing conditions in an analog condition, and 2) terminological issues that arise with regard to the manner in which to speak of two (or more) types of potential functions utilizing behavior analytic terms that are precise and technical, and that addresses different reinforcement histories. We have proposed some conventions for terminological usage to refer to two different types of function: "original function" and "current function." We have also proposed a modification to the methodology in the form of a functional assessment approach which can quickly identify changes in EO effects on reinforcement and which aids in identifying "original function" by conducting pointed, brief analog conditions, conducting a molecular analysis of the data during the actual analog conditions, and then moving on to evaluating effective treatment interventions. The suggestions we are making are intended to stimulate further discussion and hopefully research, particularly on the problem of identifying the effects of prior histories of reinforcement (Wanchisen & Tatham, 1998). If one can determine a means of assessing the effects of prior histories of reinforcement, one has a better opportunity to identify the original function.

To further clarify our approach, we are proposing that identification of the original function of behavior is crucial to more effective and efficient treatment, and to do so, one must determine or eliminate current function. We further propose that behavior maintained by a current function is weaker than that maintained by the original function that was responsible for the emergence of the behavior to begin with. If we eliminate the original function of a target behavior, current function(s) may continue to maintain it, but the behavior will be weaker and much less stable. The reason for this is that current function serves to support behavior, but eliminating the current function is not sufficient to reduce or eliminate rates of the target behavior. If we can find the original function, which is primarily responsible for maintaining the behavior, it will lead to a more effective and efficient treatment intervention. In application, we would expect that if we were to eliminate or alter the effects of current function(s), we may see either a small decrement in rate or frequency of the behavior, or a transient effect consisting of an extinction burst with stabilization at or near the previous rate or frequency of behavior. However, if we eliminate or alter the effects of the original function, we would be more likely to dramatically reduce or eliminate the rate or frequency of behavior very rapidly.

In our case example discussed earlier, the approach assists in confirming the identification of the original function while also evaluating treatment effectiveness to distinguish between current function, teacher attention, and original function, discomfort created by the ointment in the student's ears. Consequently, we can see that eliminating teacher attention may serve to weaken or reduce rates of behavior to some extent. However, because we have not identified the original function (ointment in the ears resulting in discomfort) we will not be able to finish our task of completely eliminating the target behavior, and we will not be efficient in determining the appropriate replacement behavior or effective treatment intervention.

Benefits of a Molecular Approach to Functional Assessment

The molecular functional assessment approach may have several benefits that address these concerns, since it looks at the behavior relative to function in very short periods of time. These benefits will be listed and discussed below.

1) As suggested by the name we chose, the molecular approach assesses function on a moment-to-moment basis (in this study, we look at the progression of behavior in 30-second segments within a given analog condition). Standard methods tend to look at more molar measures such as rates or averages which can make it difficult to evaluate effects of extinction or determine whether a shaping process is occurring. Such analyses are crucial to distinguishing current vs. original function.

2) Each of the analog conditions are alternated with strategic environmental manipulations to determine the original function. The molecular approach will minimize the problem of establishing a new (current) function, and help assess the existence of a current function. By evaluating data on a moment-to-moment basis, the behavior analyst is better able to distinguish patterns suggesting that the behavior being evaluated is resistant to extinction or extinguishes rapidly. When a reinforcer is presented contingently in the initial analog condition that is presumed to be function of the behavior, and in subsequently is withdrawn, a molecular analysis of the data will reveal one of two primary patterns. First, resistance to extinction with an extinction burst would suggest that the behavior was more likely to have been in the person's existing behavioral repertoire (original function). However, if the behavior extinguishes rapidly in this subsequent condition, the conclusion may be made that behavior that was maintained by the reinforcer in the prior analog condition was weak, was more recently shaped, and that one has determined a current function.

4) Another potential benefit of our approach is that one may more readily observe effects of altering EOs via providing the reinforcer of which the behavior is a function (i.e. original function). Frequently, continued presentation of the reinforcer initially responsible for the behavior (original function) will both strengthen the future probability of the behavior on which it is contingent, and may also temporarily result in immediate cessation of the behavior due to the altering effect of the EO for the behavior. However, the behavior being evaluated may not re-occur again during a brief functional analysis period due to absence or alteration of the EO for that behavior. Our approach may offer a solution to this problem by evaluating rapid changes in behavior during the course of the analysis at a molecular level (i.e. moment to moment, or minute to minute). For instance, if a reinforcer is provided contingently, and we see the behavior occur initially and then cease, one may conclude that the EO has been altered, i.e. value of the related reinforcer has been altered (reduced in this example). 4) An additional benefit of our approach is that it offers an evaluation of effective treatment intervention, which further supports the conclusion regarding the hypothesis statement for challenging behavior(s) being evaluated. This is accomplished by first determining original function, then proceeding within the same session to another condition in which replacement behaviors linked to the presumed original function are implemented in another analog condition. For example, in our experiment, Jimmy was already able to mand verbally for food by saying "eat" and had another functional skill of participating in table-top tasks independently, such as completing puzzles. The experimenters provided him with a positive activity, which he readily engaged in, and appeared to have reinforcing properties as a means of teaching him to wait short periods to gain access to food items. Hence, we were able to implement an ecological manipulation without requiring teaching any additional functionally equivalent replacement behaviors to both eliminate aggressive behavior and reduce the frequency of manding for food. The rapid reduction in rates of the challenging behavior under study as demonstrated in the molecular analysis further supported our hypothesis regarding the original function of the behavior and also revealed an effective, non-restrictive treatment intervention for Jimmy.

Conclusion

In this paper, we have presented a novel conceptualization of functional assessment in which we propose that potential functions that have been identified may not be the reinforcer responsible for the original shaping of the behavior being targeted for assessment and treatment. We present a theoretical position as to how the behavior may come to be maintained in part by other reinforcers that entered into the situation at a later time, including during functional analysis conditions. We present a unique methodological approach to resolving this potential problem, and data from a study we conducted utilizing our molecular functional analysis approach. In this approach we distinguish original function from current function based upon knowledge from the literature about prior history of reinforcement and maintenance of behavior (Tatham & Wanchisen, 1998), and the analog conditions that we conducted. We propose that it becomes crucial to identify the original function of the behavior for treatment to be effective and complete. If one does not identify the original function, there is a real risk of teaching what merely appears to be a functionally equivalent replacement skill, or implementing an inappropriate intervention (e.g. medical, ecological change). However, when the EO that evoked the behavior that was part of the original function re-occurs, then the behavior will re-occur. As a result, our data may lead us to erroneous conclusions regarding function and the appropriate treatment intervention. When our treatment fails, then we need to conduct further assessments to determine why our intervention was not effective. This state of affairs would be compromising from an ethical perspective and a financial or managed care environment in which clinical services and treatment need to be efficient and effective. Our preliminary research offers promise for this approach to determining function, and may lead to further developments as additional studies utilizing the method are conducted and yield data bearing upon our discussion of functional assessment methodology.

REFERENCES

Barlow, D.H., & Hayes , S.C. (1979). Alternating Treatments design: One strategy for comparing the effects of two treatments in a single behavior. Journal of Applied Behavior Analysis, 12, 199-210.

Hersen, M., & Barlow, D.H. (1976). Single case experimental designs: Strategies for studying behavior change. New York: Pergamon Press.

Horner, R., & Dodson, S. (1993). Manipulating setting events to decrease problem behavior: A case study. Teaching Exceptional Children, 25, 53-55.

Iwata, B.A., Dorsey, M.F., Slifer, K.J., Bauman, K.E., & Richman, G.S. (1994). Toward a functional analysis of self-injury. Journal of Applied Behavior Analysis, 27, 197-209. (Reprinted from Analysis and Intervention in Developmental Disabilities, 2, 3-20, 1982).

Kennedy, C.H., Meyer, K.A., Knowles, T., & Shukla, S. (2000). Analyzing the multiple functions of stereotypical behavior for students with autism: Implications for assessment and treatment. Journal of Applied Behavior Analysis, 33, 559-571.

Mace, F.C. (1994). The significance and future of functional analysis methodologies. Journal of Applied Behavior Analysis, 27, 393-400.

Mace, F. C., & Knight, D. (1986). Functional analysis and treatment of severe pica. Journal of Applied Behavior Analysis, 19, 411-416.

Magee, S.K., & Ellis, J. (2000). Extinction effects during the assessment of multiple problem behaviors. Journal of Applied Behavior Analysis, 33, 313-316.

Michael, J. (1993). Establishing Operations. Concepts and principles of behavior analysis. Kalamazoo, MI: Society for the Advancement of Behavior Analysis.

O'Neill, R.E., Horner, R.H., Albin R.W., Storey, K. & Sprague, J.R. (1997b). Functional analysis of problem behavior: A practical assessment guide. Sycamore, IL: Sycamore.

Schillinger, H., & Blakely, E. (1994). A descriptive taxonomy of environmental operations and its implications for behavior analysis. The Behavior Analyst, 17, 43-57.

Tatham, T. A., & Wanchisen, B. A. (1998). Behavioral history: A definition and some common findings from two areas of research. The Behavior Analyst, 21, 241-251

Ulman, J.D., & Sulzer-Azaroff, B. (1975). Multielement baseline design in educational research. In E. Ramp & G. Semb (Eds.), Behavior analysis: Areas of research and application (pp. 371-391). Englewood Cliffs, NJ: Prentice-Hall.

Vollmer, T.R., & Smith, Richard G. (1996). Some Current Themes in Functional Analysis Research. Research in Developmental Disabilities. 17,3, 229-49.

Vollmer, T.R., & Vorndran, C. M. (1998). Assessment of Self-Injurious Behavior Maintained by Access to Self-Restraint Materials. Journal of Applied Behavior Analysis, 31, 647-650.

Wanchisen, B. A., & Tatham, T. A. (1991). Behavioral history: A promising challenge in explaining and controlling human operant behavior. The Behavior Analyst, 14, 139-144

(1) This article is a reprint. It was previously published in Behavior Analyst Today, Volume 2, No. 3, September, 2001.

WILLIAM MARSH, M.S., BCBA, MICHAEL WEINBERG, PH.D., BCBA, AND ANDREW HOUVOURAS, M.A., BCBA

DEVEREUX FLORIDA CAMPUS PROGRAMS, VIERA, FLORIDA
COPYRIGHT 2005 Behavior Analyst Online
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2005 Gale, Cengage Learning. All rights reserved.

 Reader Opinion

Title:

Comment:



 

Article Details
Printer friendly Cite/link Email Feedback
Author:Marsh, William; Weinberg, Michael; Houvouras, Andrew
Publication:The Behavior Analyst Today
Date:Mar 22, 2005
Words:6251
Previous Article:Near-infrared spectrometry for the measurement of central nervous system activation: a brief demonstration of an emerging behavioral assessment tool.
Next Article:The role of verbal conditioning in third generation behavior therapy.


Related Articles
MEMOREC Stoffel GmbH Licenses Genzyme's SAGE Technology.
Position Announcements.
MIRADA/HITACHI LAUNCH NEW EDITION OF FUSION7D FOR AVIA.
Research and Markets: Building Brand Equity in the Molecular Biology Market - Over 1,500 Life Scientists Surveyed.
Toxicogenomics: Principles and Applications.
How do flies fly? Does projectin function as an elastic band?
Molecular biomarkers predicting efficacy of cancer therapy highlighted.
Methodological and terminological considerations of functional assessment: towards a molecular analysis of function.
New biology and Bioenergy titles from Springer.
Molecular Biometrics Hires Denny Sakkas, PhD Renowned Yale Embryologist and International Thought-Leader, as Chief Scientific Officer; Opening U.S....

Terms of use | Copyright © 2014 Farlex, Inc. | Feedback | For webmasters