It is estimated that for every 1,000 children born, two to six will be diagnosed with autism. Because of its relatively large presence, treatments and therapies for the impairments autistic children face are constantly being developed and improved. One of the most significant impairments associated with Autism is extreme difficulty in the development of speech and language. Some of the features of the language and speech impairment that autistic children face are unusual word choice, pronoun reversal, incoherent discourse, unresponsiveness to questions, and a lack of drive to communicate among other problems. Despite these problems, over half of children diagnosed with autism possess the abilities to understand and produce speech and eventually develop a certain degree of functional speech. Music therapy has been studied as a potentially effective way to teach autistic children functional vocabulary words, however the previous concerning the effectiveness of music therapy studies have not been sufficiently controlled.

So, to study the effectiveness of music therapy in a more controlled situation, Lim (2010) studied whether or not learning functional words through music stories was more effective than learning words through regular speech therapy type stories. A total of 50 autistic children between the ages of 3 and 5 were used in this study. Based on the Childhood Autism Rating Scale or the Autism Diagnostic Interview Revised, the children were categorized as either ‘low-functioning’ or ‘high-functioning’.  The target words for this study were selected from lists of functional vocabulary words that 3-year-olds could use in everyday interactions.

Knowledge of the 36 target words was evaluated with pre- and post-tests of the children. The tests took the form of a verbal fill-in-the-blank test where the phrase before the blank cues the correct word without being similar to the word. For example, the phrase: “Old MacDonald had a _____” cues the word “farm” without mentioning or being similar to the word “farm”. All of the phrases in the test were structured to end in the target word, like the example above. A picture of the target word was also presented to help prompt the participant to produce the target word upon hearing the first part of the phrase. It is important to note that a different picture was used every time (different types of farms, etc) to demonstrate that the children were not simply memorizing the picture.

The children were randomly placed into three groups: the music therapy condition, the speech therapy condition, and a control condition.  For the music therapy condition, six songs were composed that contained six target words each encompassed all of the 36 target words. Each line of the song ended with a target word, for example:

Hello, hello brown bear.

What do you eat?

I’d like to eat an apple.

After eating apples, he is happy.

Brown bear says more.

Daddy bear says the apples are all gone.

Each song varied by key (i.e. C Major, d minor, etc), tempo (the speed at which the song is being played), and meter (the rhythm) but they were all similar in that they were very simple and repetitive songs, as one could imagine from the lyrics. In each song, the target word, which was also the last word, was always emphasized by a longer time duration (i.e. Daddy bear says the apples are allllllll gonnneeeee). There was also a picture presented in the video by the singer with every target word (but as in the pre- and post- tests, the same picture was never shown twice). All of the songs were sung by a female music student with guitar accompaniment. The songs were videotaped to be shown to participants on a TV monitor with each song being repeated twice in each video.

For the speech training condition, the same text from the six songs was simply recited without any musical accompaniment or melody. As in the music therapy condition, a picture was presented with each target word, the same female music student recited the stories, and the stories were videotaped and repeated twice each to show the children.

Finally, for the control condition, no training was given.

They were first given the pre-test (the verbal fill-in-the-blank test described above), then were either shown the music video, speech video, or no video (control condition). Each participant saw their respective video twice a day (which totaled four exposures to the song/speech a day since each song/speech was repeated twice in the video), for three days (totaling 12 exposures to the song/speech). On the fourth day, the post-test was given to each child. All of the pre-tests and post-tests were videotaped. It is also important to note that the pre-test, the video sessions, and the post-test were given to children individually to prevent the child from being distracted.

Each response given by each child was evaluated by two speech pathologists who assigned points to each response based on semantics, phonology, pragmatics, and prosody.

Semantics refers to if the child produced the correct target word.

Phonology refers to the correct pronunciation of the target word.

Pragmatics refers to the time delay between being presented with the verbal blank and the response of the target word.

Prosody refers to the way the word is said—which syllable is emphasized, question vs statement, etc.

Various levels of all of these criteria were assigned point values and then the speech pathologists scored each pre- and post-test video of each child. Both of their scorings were almost identical which demonstrates consistency in the rating system and the raters.

It was found that both the music and speech therapy conditions produced significantly more improvement in the verbal production score than the control condition. Overall, it appeared that there wasn’t a significant difference between the effectiveness of the two types of therapy. However, upon further investigation, it was found that the music therapy produced significantly more improvement with ‘low-functioning’ children than the speech therapy, while the ‘high-functioning’ children showed equal improvement in both the speech and music therapy conditions.

The graph shows the differences in scores between the pretest and posttest. The bigger the number, the bigger the improvement in speech production.

Though the speech and music therapies only differed when comparing low- and high-functioning children, this finding is significant. It suggests that the verbal instructions may be too confusing or the speech task not exciting enough to hold the low-functioning children’s attention. During this critical age for children to learn speech, it is especially important for the low-functioning children to be exposed to a therapy that is effective for them and this appears to be music therapy. The author of the study suggests that the predictable temporal pattern of music makes it easier to perceive by the low-functioning children and that the “exciting” nature of music holds the attention of low-functioning children better. Based on these findings, verbal therapies for preschool autistic children should heavily include music therapy since it is significantly effective for both the high- and low-functioning children. With research like this, we can improve the chances that an autistic child can grow up to become an independent, functional member of society instead of being dependent on others.

Lim HA (2010). Effect of “developmental speech and language training through music” on speech production in children with autism spectrum disorders. Journal of music therapy, 47 (1), 2-26 PMID: 20635521

Have you ever heard a song on the radio that you recognize but you can’t pinpoint the name of the song? From personal experience, I know this happens to me a lot! I can hear a song on the radio and know I have heard it before but I cannot recall the name the song. It is common for people to recognize a song but how much or how little of the song do you need to hear to recognize it? Do you think you could recognize a song if you just heard a fragment of the song, the rhythm or if you just heard the notes played for you without any rhythmic information? Kostic and Clearly (2009) wanted to explore just that.

Kostic and Clearly conducted four different experiments to understand what parts of a song are essential for you to remember hearing it. Most of the participants in their study had musical experience. The researchers had participants listen to known song clips that captured the main theme of either a children’s melody, popular rock song or a pop song. The melody of the songs was played by a piano. These clips were used for the study session of the experiment. The participants were asked to identify the song after the clip was played in the study session. After this identification, the participants were told the name of the song to ensure that they had some knowledge of the song’s name within the context of the experiment. The participants were told that they would have to recall the song’s name for a later memory test.

Try this out for yourself. The song clips in this video are the length of those heard in the study session but these clips include all of the information from a song where as the participants in this study only heard the melody line.

How did you do? My guess is you recognized most of the songs but it was difficult for you to name the tune.

Up next was a test session. In this session, participants were presented song fragments that were distorted but still maintained pitch and rhythmic information. The researchers wanted to see if a change in quality of the song’s sound would alter recall. They asked participants to name the song that they had just heard. After this the participants were asked to rate the likelihood that they heard that song in the study block. Half of the songs in the test block were heard in the study session and half were not. This part of the study yielded the expected results. Participants were able to recognize a song but were not able to name the tune.

The researchers did three variations on the above stimuli with new groups of participants for the next three experiments. The task was exactly the same as the first experiment but they changed what the participants heard in the test session. The researchers wanted to know if this effect would still be seen if participants only heard a certain component of the music or if this effect was only possible when all of the musical information was intact.

In the second experiment, the researchers had the participants listen to songs that only had pitch information (i.e. without any rhythm) during the test block. They set each song to a standardized rhythm that made it impossible to recognize a song as familiar purely by its rhythm. In this variation, they had two conditions for note order: scrambled and ordered. This allowed the researchers to study whether note order is important to hear a song as a familiar. The results for this variation found that participants were able to recognize a song as familiar but were not able to identify it. However, when the notes were in a scrambled order participants had a hard time recognizing the song as one they had heard in the study session. This implies that while you can recognize a song based solely on the notes played you need to hear them in their original order to recognize it as familiar.

During the third experiment, participants heard the songs played on a wood block which made the participants rely on the rhythmic information when deciding if they had heard the song in the study session. Once again, they found that participants were able to recognize a song as familiar despite only hearing the rhythm of the song. However the participants were still not able to name the songs.

The last variation was on the tempo or speed of the song. If the song was originally slow, when heard in the study session, they sped up the tempo and slowed the tempo down for fast songs. By doing this alteration, the researchers were able to determine if tempo has to be exactly like the original song for it to be recognized or if it can be relatively the same. The results illustrated that tempo of a song can be changed and people still recognize it as familiar. But, they still could not name the song.

The findings from this study reveal that there really is not just one aspect of a song that makes you recognize it as familiar. You can hear a distorted fragment of a song and recognize it. You can hear just the notes of the song played in order and recall that you heard the song before. You can hear the rhythm of the song and still remember it. Finally, you could hear the notes of a song played at a different speed and still recognize it. This speaks to the variety of things we attend to when we are listening to a song. The next time you are listening to a song and cannot recall its name do not feel ashamed it happens to the best of us.

Kostic B, & Cleary AM (2009). Song recognition without identification: when people cannot “name that tune” but can recognize it as familiar. Journal of experimental psychology. General, 138 (1), 146-59 PMID: 19203174

Have you wondered if the child in the backseat is noticing the different songs that have come on the radio?  Or have you ever been listening to NPR and the music suddenly changes from a classical Western piece to an instrumental piece of an unfamiliar type?  Now, would the child in the backseat notice such a song change?   A recent study shows that children and adults do not differ in a memory task identifying music selections based on music culture type (familiar or unfamiliar).  Their findings point to enculturation and the possible effects it may have on memory for music.

Enculturation is the process by which a person achieves a cultural specific understanding for something that is perceived.  Morrison and his fellow researchers set out to explore the role of culture and its effect on memory for music under different complexity conditions for two different age groups.  This study addresses musical enculturation specifically.  Musical enculturation is the formation of music schemata, or the mechanisms by which a person interprets music that is perceived.  Many anthropologists and music specialists have investigated the effect of culture on music understanding.  These researchers aimed to extend the findings of a previous study by Demorest, Morrison, Beken, & Jungbluth (2008), which investigated the effects of familiar and unfamiliar music culture on music memory for trained and untrained listeners.   Studies have found differences between Western and non-Western participants for aspects of music such as melodic perception and emotional response.  A challenge to this type of research lies in defining just what is musical understanding as well as whose judgment is the right one.  Studies have shown participants to be better at tasks that are predictable and logical to them.  Based on these findings, we would expect to see cultural-specific patterns in response to music that is predictable and logical.

The purposes for this study were twofold.  The first purpose of this study was to examine the effects of simple and complex musical structures for familiar and unfamiliar cultures.  The second purpose was to investigate the differences between children and adults for the conditions for memory for music mentioned above.  The specific research questions were:

1. Would memory performance for children and adults be better for their home culture (Western music) than for an unfamiliar culture (Turkish music)?
2. Would enculturation effects vary, depending on the complexity of the music stimuli?
3. Would there be any overall difference between the performance of adults and children on the memory task?

The 90 participants in this study were fifth-grade students (n = 43) and college-age adults (n = 50).  The average age of each age group was not reported.  All participants were born and raised in the United States.  There was a questionnaire to identify the primary culture in which they were raised.  The questionnaire did not include questions about musical experience.

There were two sets of stimuli: simple and complex, one of each for Western music (familiar culture) and Turkish classical music (unfamiliar culture).  The researchers chose Turkish music due to its use in previous research, as well as distinctive musical qualities (e.g., timbre and mode).  None of the musical selections included lyrics.

The complex stimuli for this study were identical to those in Demorest et al. (2008), the research the current study is extending.   The complex stimuli included “wind and string instruments and with minimal internal repetition within each excerpt” for both Western and Turkish musical selections (Demorest et al., 2008).  The complex stimuli for the unfamiliar condition were selected by trained musicians who had studied Western and Turkish music.  They selected Turkish pieces based on the Western selections to ensure the closest match (culturally) for each condition.

The simple conditions for the Western music condition included a classical guitar solo.  A similar Turkish instrument, the ud, was used for the simple Turkish music condition.  To avoid priming effects and cultural biases, experts approved the musical selections as obscure and not likely to have been performed or heard.  The examples were about 30 seconds (s) in length.  Following the background questionnaire, the participants completed a familiarization task involving a short jazz selection and 2 memory test items.  The short jazz selection included one target and one foil, which served as the test items.  A target was a short bit of music included in the jazz selection that they had heard before (“yes” being the correct response).  A foil was a short bit of music not included in the jazz selection (“no” being the correct response).  If they had no questions after the familiarization task, the participants proceeded to the experimental tasks.

The participants were randomly assigned to the simple or complex condition.  There were 46 participants in the simple condition and 47 in the complex condition.  They do not report how many participants from each age group were included in the simple and complex groups.  The musical selections were played on a portable CD player. For each testing procedure, the researchers told the participants to listen carefully as 3 longer selections were played, followed by a memory test of 12 items.  The memory test included 6 foils and 6 targets, all ranging from 4 to 8 seconds long.  The foils and target were taken from musically different selections of the same pieces.  The aim of foils and targets, therefore, was to force participants to make their decisions for recognition based on the structural aspects of the music (e.g., melody and harmony) and not the low-level aspects of the music (e.g., recording quality, number of instruments).  They were allowed 5 s to answer the “yes/no” question following the presentation of each foil or target.  The foils and targets were mixed for culture and randomly presented.  The presentation order was the same across participants. The participants were presented with the Western music clips first, followed by the test items for this portion of the task.  The Turkish music selections were then presented, followed by test the items.  There may be concern for priming effects for culturally familiar music.  However, Demorest et al. (2008; cited in Morrison et al., 2008) found presentation order of culturally familiar and unfamiliar stimuli to have no effect on a memory task.

The researchers found a significant difference between groups for music culture.  The U.S.-born children and adults were significantly better at the memory performance task for novel music from their home culture.  The groups showed significant difference in performance for music complexity.  Overall, the participants demonstrated that the simple selections were easier to remember than the complex ones.   The only significance between children adults was observe din the complex-familiar condition.  The adults did not, however, perform this well in the complex-unfamiliar condition.  The latter finding suggests that memory related to complexity may not be a matter of age, but of cultural familiarity.  As you can see in Figure 1, children appear to perform better than adults in the complex-unfamiliar task.  However, this difference was not significant.

*Please note that memory scores for all participants were coded into d-prime (d’), which is a statistic used to standardize memory task data.  This statistic takes into account response bias by measuring the ratio of correctly identified targets to falsely identified foils.  For the sake of this blog, trends and significant differences (noted above) are more important than the specific d’ values.

Morrison, Demorest, & Stambaugh (2008) set out to investigate the difference between adults and children for a memory task involving culturally familiar and unfamiliar music.  They also included a complexity manipulation to further explore the effects of musical complexity on the memory task.  As stated above, the researchers found significant differences for the memory task for music culture.  Both children and adults scored significantly higher on the culturally familiar (Western) music than the culturally unfamiliar (Turkish) music.  These results tell us that a cultural bias for music memory happens sometime before adulthood.

Based on Morrison et al.’s findings, we can say that there is something about culture that affects our ability to recognize music.  This musical enculturation makes it easier for us to recognize simple and complex music selections that are familiar better than even simple unfamiliar ones.  Children and adults are equally good at this, except in the complex-familiar condition.  The researchers propose that this difference may be due to a “richer knowledge” of culturally familiar music.  Keep these findings in mind next time you see children dancing to seemingly unfamiliar music.  They may enjoy the rhythm and the instruments they hear.  If asked, however, they may identify it as equally unfamiliar as you do!

Morrison, S., Demorest, S., & Stambaugh, L. (2008). Enculturation Effects in Music Cognition: The Role of Age and Music Complexity Journal of Research in Music Education, 56 (2), 118-129 DOI: 10.1177/0022429408322854