Effects of music on task performance,engagement,and behavior: A literature review

Music in relation to task performance and task engagement

Active engagement with academic or vocational tasks results in better performance and outcomes (Iovanne, Dunlap, Huber, & Kincaid, 2003). Eight studies in this review discussed the facilitating effect of music on increasing task performance and engagement in individuals with developmental disabilities. These studies analyzed the effects of different types of background music on the performance of various tasks.

In comparison to listening to white and pink noise (sounds with all frequencies often used to block distractions and sounds with high and low frequencies for relaxation purposes, respectively), individually selected music (either classical, ballad, western, disco-rock, or children’s songs) produced the greatest work output riding a stationary bicycle as calculated by an ergometer for 10 out of 13 participants (Caouette & Reid, 1991). Four of the 10 participants increased their work output by nearly 2000 watts over 15 minutes. Researchers equated all tempos (112–124 BPM) to control for possible tempo influences on their bicycle performance. Similar results illustrating a significant difference using instrumental background music versus no music were evident in Stainback et al. (1973) with relevant daily living tasks. Additionally, Burleson, Center, and Reeves (1989) and Groeneweg et al. (1988) evaluated the effects of instrumental music on work accuracy using withdrawal designs. Both studies compared participants’ task performance in a music and no-music condition in order to determine the frequency of work-oriented and productive behavior versus non-work-oriented behavior (i.e., off-task behavior) on academic and vocational tasks respectively. Burleson et al. (1989) examined the effects of instrumental background music on the accuracy of a known color-coded sorting task while Groeneweg et al. (1988) assessed individuals sorting grocery-store coupons. Both studies illustrated an increase in task accuracy and a decrease in non-work related behaviors when music was playing. Overall, productivity was highest when background music played, which resulted in a significant and positive effect on the task performance. Music may be effective in stimulating a person to respond to simple and repetitive tasks rather than complex tasks. Because these four studies used somewhat predictable, structured tasks with repetitive activities, the effects of background music on task performance may be task specific (Groeneweg et al., 1988).

Music playing at a regular tempo also showed the greatest increase in the number of letter envelopes filled (mean n = 31.12) by the participants when compared to fast tempo (mean n = 28.00), slow tempo (mean n = 28.71), and no music (mean n = 23.36) at a sheltered workshop (Richman, 1976). The noncontingent application of background music in this study showed an improvement in task performance, though this was not statistically significant, again supporting the hypothesis that background music can increase task performance in persons with intellectual disabilities. Possible explanations for this finding could be that the faster tempo overstimulated the participants, resulting in less body control whereas the slower tempo decreased activity levels to a degree which resulted in poor attention. This was the only study which investigated how manipulating the tempos in music influenced task performance, indicating that musical tempo and tasks could be matched to facilitate performance.

A more recent study by Carnahan, Musti-Rao, and Bailey (2009) took the role of music in task engagement a step further and paired it with visual materials in order to determine whether the combination of two strategies would increase the academic engagement of six elementary students with ASD. Engagement included (a) students orienting their eyes and body towards the learning materials, (b) using the learning materials appropriately, (c) making verbal responses or comments during instruction related to the materials, and (d) the absence of self-stimulatory behaviors. The authors adapted commercially written books and teacher-written books to include interactive pieces such as laminated photos and three-dimensional materials that could be manipulated by the students. The text from the books was sung with guitar or piano background music being played during some phases of the study and only read aloud by the teacher in other phases. When the interactive materials were used with music, participant engagement increased by 51–91%. However, when interactive materials without music were used, engagement for some students was decreased. Overall, the results showed that both types of stimulation were necessary to increase the academic engagement of students with ASD. In addition to the participants’ increased engagement, the teacher implementing the strategies found the materials to be useful, motivating for the students, and easy to execute.

Task engagement illustrating on-task behavior can be challenging for persons with developmental disabilities who have difficulties with maintaining attention, processing information, and communicating (Beirne-Smith, Ittenbach, & Kim, 2006). In addition, educators and practitioners often struggle to enhance the engagement and performance of persons with disabilities as they may be perceived as less capable or unwilling to learn (Carnahan et al., 2009). As a result, these individuals may be excluded from potential learning activities that could enhance their knowledge and independence due to their difficulties maintaining active attention. The studies reviewed above suggest that music can have a stimulating effect, increasing both task engagement and performance across a variety of academic, vocational, and daily living tasks. In order to ensure better outcomes in the future for individuals with disabilities, further research on strategies to increase engagement and performance must be conducted and evaluated.

Music in relation to vocal stereotypy

Stereotypic behavior can be defined as repetitive or ritualistic vocal or motor responses which seem to serve no obvious function for the individual (Ahearn, Clark, MacDonald, & Chung, 2007). This category of behavior is often presumed to be maintained by automatic reinforcement, meaning the vocalizations or movements are automatically reinforced by producing the behavior. Although there is a wealth of literature regarding stereotypy, only six studies were identified in this review that specifically focused on vocal stereotypy. Lanovaz et al. (2011) defined vocal stereotypy as “acontexual audible sounds or words produced by the vocal apparatus (e.g., tongue, lips, nasal cavity, and vocal cords)” (p. 648). This operational definition was similar in all six studies reviewed. Vocal stereotypy appears to have no communicative intent and can vary in intensity and comprehensibility (Ahearn et al., 2007).

Music presents an interesting possibility to compete with or replace vocal stereotypy in individuals with developmental disabilities. The six studies investigating the use of music to reduce vocal stereotypy in persons with ASD and intellectual disabilities in this review can be broken down into three different categories: playing background music, comparing matched and unmatched auditory stimuli, and changing the volume levels of the music.

First, research has been conducted to examine the effects of solely playing background music in hopes of decreasing vocal stereotypy while engaged in different tasks. Gunter and Fox (1993) investigated the effects of background music by applying music contingent (i.e., music was stopped when off task) and noncontingent (i.e., music played no matter the behavior) on the repetitive vocalizations and task performance of a 14-year-old male with ASD. The participant engaged in an academic task which consisted of word identification and a vocational task involving a three-step process of collating three brochures, placing them in an envelope, and finally placing the envelope on a tray. Results indicated that a decrease in aberrant behavior occurred in both the noncontingent and contingent applications of music. Interestingly, the contingent application of music resulted in improved task performance on the vocational task but not on the academic task. One possible explanation may be because the individual was less engaged in reading the sight words, which resulted in the music stopping for longer periods. Saylor et al. (2012) also found that listening to preferred songs noncontingently decreased vocalizations (near zero levels), more so than listening to white noise or the participants’ recorded vocal stereotypy while in various settings. The reduction in the repetitive vocalizations of the participant in response to both contingent and noncontingent applications of music suggests that music may serve as an abolishing operation for immediate vocal stereotypy. One possible limitation of these studies is the lack of a functional assessment to determine the function of the vocalizations. Instead, the authors implemented the above intervention in part to determine whether the behavior was automatically reinforcing. By doing so, the authors used the experimental manipulation as a partial functional analysis. Although vocal stereotypy is presumed to have sensory consequences as auditory stimulation (Patel, Carr, Kim, Robles, & Eastridge, 2000), without a proper functional analysis it is difficult to assume that music served the same function for the participant as the vocalizations. Results from both studies found music to be the most effective intervention to change behavior, while other forms of auditory stimulation, such as white and pink noise, were ineffective in significantly changing behavior.

Second, two studies evaluated the effects of music as a matched stimulus on immediate and subsequent vocal stereotypy (Lanovaz et al., 2009; Rapp, 2007). Matched stimuli are items or activities that produce stimulation similar to that of the behavior, which in this case is the vocal stereotypy (e.g., musical keyboard and radio), whereas unmatched stimuli provide dissimilar stimulation (e.g., magnetic letter boards and toys that do not make noise). Both studies found that matched auditory stimulation decreased immediate vocal stereotypy; however, the results lacked a clear abolishing operation on subsequent vocal stereotypy engagement when the music stopped playing.

Third, two studies examined the effects of manipulating the volume of the music on vocal stereotypy. Cunningham (1986) altered the volume of the background music to determine the effect on the frequency of vocalizations. Each participant with a severe intellectual disability was observed for 20-minute periods listening to instrumental jazz-rock at different intensity levels. Results showed that the loudness of the music could control the frequency of vocalizations; specifically loud music decreased the number of vocalizations by 37.5% whereas no music or soft music increased vocalizations by 35.5% and 69.5% respectively. This study was limited due to the small number of intervention sessions. Likewise, Lanovaz et al. (2011) alternated the music between high levels of intensity (i.e., loud) and low levels of intensity (i.e., soft) for two participants with ASD. Music was applied noncontingent on behavior throughout the sessions. The authors found that the immediate level of vocal stereotypy decreased with high and low-intensity songs. However, the patterns of behavior between the two participants differed over time, suggesting that more investigations are warranted. Although the results of these two studies varied slightly in their effects, both studies found a decrease in vocal stereotypy when background music was used.

The aforementioned studies focused on the efficacy of various forms of background music to decrease the vocal stereotypy of individuals with disabilities, which demonstrated the potential of music as a promising intervention for this population. One hypothesis for the effectiveness of these interventions could be that music competed with or substituted the reinforcement produced by the vocal stereotypy of the participants (Saylor et al., 2012). If the vocalizations served as automatic reinforcement for the participants, the substitution of music to fill that sensory need would account for decreases in vocalizations. However, if automatic reinforcement was not the function of the vocal stereotypy, then the ability of music to decrease the behavior requires further exploration. A functional analysis of the vocal stereotypy would have been helpful in determining the function of the behavior. Possible explanations for these results could be the novelty of the music, the participant’s preference for the type of music, or the length of the study. If the experiments were conducted for longer periods of time, perhaps the participants would satiate on the music and revert to producing auditory stimulation themselves. This line of research is in the initial stages of providing needed information to support a possible method for reducing vocal stereotypy in individuals with developmental disabilities. In addition, more research is needed to evaluate whether or not these vocalizations are distracting for others and whether they hinder social relationships and task performance.

Music in relation to aberrant and self-injurious behavior

In addition to reduction in vocal stereotypy, research has been conducted on the effects of music on aberrant and SIB of persons with developmental disabilities. Reducing self-injury and other aberrant behaviors is a challenge facing those who work with this population (Ford, 1999). Not only can these behaviors cause physical harm, they can prevent the acquisition of other important skills. The need to reduce these behaviors is irrefutable, yet often the answer is the use of restrictive and punitive measures such as medication, physical restraints, electric shock, or solitary confinement (Davis, Wieseler, & Hanzel, 1983; Wells & Axe, 2013) with few opportunities for integration into the community. Music is one alternative as a more positive and less intrusive intervention to reduce the frequency and intensity of these behaviors but refrain from restraining or punishing the individuals.

Rumination is one example of an aberrant behavior. It is when food is regurgitated and then either rechewed or spit out, often resulting in malnutrition, damage to the teeth and mouth, and even death. Davis et al. (1983) proposed examining the effects of the contingent removal of music paired with a verbal reprimand of “No” as a less aversive intervention to reduce rumination as well as out-of-seat behavior. The researchers found that contingent music alone was effective in reducing out-of-seat behavior and, when used concurrently with the verbal reprimand “No”, was effective in reducing rumination to nearly 0%. Without the verbal reprimand, rumination was decreased by 75% with the contingent removal of music alone. The results indicated that music could be a non-intrusive method to reduce maladaptive behaviors such as rumination.

In another study, Ford (1999) investigated the use of less restrictive methods for reducing head-hitting, teeth-grinding, and mouth-scratching for an adult woman with a severe developmental disability. Ford (1999) proposed that the SIB could be “an external communication that an internal need has not been met in the environment as it is” (p. 294). Therefore, she hypothesized that music could perhaps satisfy that internal need instead of injuring oneself. She compared music to other highly reinforcing stimuli (i.e., swimming and instrument playing) to determine whether there was a difference in the rate of SIB following the different conditions. Results showed that listening to music resulted in a slight reduction of teeth-grinding; however, there was not a significant change in other behaviors. Likewise, Orr, Smith Myles, and Carlson (1998) saw a slight decrease in the frequency of head jerks and screaming when rhythmic music was played for an 11-year-old female with ASD. The lack of a functional analysis illustrating the function of the SIB (Ford, 1999) or the inconclusive results (Orr et al., 1998) in these studies may have contributed to the results showing little change in reducing the SIB. Consequently, Carey and Halle (2002) did initially conduct a functional analysis and found noncontingent background music to decrease SIB for one young adolescent with a severe intellectual disability. A teenage girl with an intellectual disability and blindness also decreased her self-stimulatory behaviors when presented with loud (slightly above conversational level) background music of her choice versus no music (Desrochers, Oshlag, & Kennelly, 2014). Similarly, another teenage girl with an intellectual disability and blindness increased her in-seat behavior when listening to rock music as opposed to no music (Hill et al., 1989).

In an earlier study, Reardon and Bell (1970) assessed the frequency of active and sedentary behaviors during three different conditions: sedative music, stimulative music, and a read-aloud story using an alternating treatments design. While the study lacked fidelity and reliability data, music was not shown to be as effective as a story being read aloud. Future research following single-subject best practice guidelines (Horner et al., 2005), which include collecting and evaluating procedural and inter-rater reliability, should be conducted to confirm or refute these results.

In the treatment of SIB and other maladaptive behaviors, these studies provide some support for the use of background music to reduce this class of behaviors. One explanation for the effectiveness of this intervention to reduce aberrant behaviors is that listening to music is reinforcement, specifically contingent reinforcement in the studies mentioned above. This would account for the decrease in behaviors that potentially disrupted or competed with access to music. Another possibility is that the music masked other auditory stimuli in the area which were potentially aversive to participants. Although these studies had mixed results, they show some promise in our understanding of what might be reinforcing or comforting to individuals with disabilities. This direction of research could not only improve behavior, but change people’s perspective on the issue of SIB. People often do not understand that SIB serves a purpose as a self-stimulatory behavior or a form of communication. Researchers should further investigate these music alternatives and report on the efficacy of their findings in order to expand the literature on less restrictive methods, such as music, to reduce self-injurious and aberrant behavior. Ethical guidelines mandate that the least intrusive methods should be employed first, and only after these methods have proved unsuccessful should more restrictive methods be considered (Cooper, Heron, & Heward, 2007).

Limitations of the studies reviewed

A limitation found in all of the studies reviewed was the failure to include a maintenance phase to determine whether the behavior change continued over time. Ideally, the increase in performance and the decrease in aberrant behaviors would continue to be evident for months and years after the initial changes and not just as the result of novel stimuli (Ford, 1999). In addition to including a maintenance phase, studies should evaluate generalization. All but three studies (Davis et al., 1983; Gunter & Fox, 1993; Saylor et al., 2012) lacked generalization measures to test whether the behavior change generalized to other environments, tasks, or behaviors. It is vital that future research investigate the durability of the behavior change by building maintenance and generalization measures into the experiment.

In addition, this literature review revealed a need for studies implemented with fidelity incorporating the quality indicators of single-subject research (Horner et al., 2005) because the majority of studies including participants with developmental disabilities have a small number of participants. Within the 44-year span of this review, 8 of the 20 identified studies were published before 1990. Earlier studies often lacked procedural definitions and sufficient data collection methods; therefore results can be interpreted as a therapeutic trend at best. This lack of procedural clarity in many of the studies presents difficulty for systematic replication, which is an issue that must be addressed in future research investigating the effects of background music on performance and behavior.

Although the changes observed in these studies support more investigation into music as a potential strategy, it is difficult to view the results as significant without sufficient data points, procedural clarification, and adequate descriptions of controls to internal validity. Without a valid research design and adequate data collection demonstrating within and between replications of effect, the effectiveness of music cannot be conclusively determined. Ford (1999) specifically discussed the lack of environmental control (i.e., changes in staffing, background noise, and behavior of others present), which weakened the interpretation of the data. Future studies should address these issues in order to extend the quantitative research on this topic and provide a strong evidence-base.

Limitations of this review

While a limitation of this literature review is the inclusion of articles that do not meet all of the quality indicators, this limitation helps to illustrate the need for additional high-quality research. A stronger research base on the effects of background music could lead to improved outcomes for individuals with developmental disabilities. However, the results of this review should be interpreted cautiously as comparisons across studies were often difficult to make given the limited information provided, especially from some of the older articles. The lack of operational definitions and clear procedures restricted the findings and ability for replication.

An additional concern when interpreting the results of this review is the limited generalizability of the findings due to the heterogeneity of participants. Participants both within and between studies differed in age, diagnosis, and severity of disability. Even though this review focused on participants with developmental disabilities, a broad continuum of individuals diagnosed with ASD or an intellectual disability exists. Another limitation is the small number of studies in each sub-category as this review specifically evaluated the effects of background music for a particular population of persons with disabilities. It is difficult to draw conclusions with so few studies and participants, but it does illustrate the need for additional research.

Future directions

Even with these limitations, the studies included in this review showed promising results and provided a rationale for continued research into the use of background music to influence behavior. The importance of identifying strategies to replace aberrant behavior and increase task performance in persons with disabilities is undeniable. A possible means to do so could be incorporating background music as an environmental modification to the setting or through iPods, MP3 players, iPads, or computers with headphones or earbuds for individuals. How music affects specific types of behavior and the duration of that behavior change across tasks and environments needs to continue to be investigated so music can be better understood and used by educators, clinicians, and families as an effective strategy. Future studies should continue to examine the effects of music on different types of academic and vocational tasks as well as the ability of music to produce lasting behavior change. Studies should also examine the effects of different types of music on these behaviors as well as the length of exposure to the music. Another aspect of background music which needs further examination is the comparison of background music as a stand-alone intervention versus its use as a component of an intervention plan. In order to determine why music can decrease vocal stereotypy for individuals with a developmental disability, future studies should (a) complete a functional analysis of the vocal stereotypy prior to intervention in order to determine the reinforcement contingencies which maintain the behavior, (b) experiment with different types of music, and (c) conduct the study for a sufficient length of time to test for the possibility of satiation.

These questions and others must be examined in order to better understand the influence of music on behavior and why it is an effective intervention for individuals with disabilities. Without continued investigation into how to increase the performance and engagement of these individuals, they may lose opportunities to learn important skills which could help them achieve more independence and success in the future. Also, if music can be a replacement behavior for vocal stereotypy, aberrant behavior, and SIB, then the time to conduct a functional analysis and implement the music strategy would be worth it.