PERCEPTION OF RHYTHM by Adults with Special Skills Annual Convention of the American Speech-Language Language-Hearing Association November 2007, Boston MA Elizabeth Hester,, PhD, CCC-SLP Carie Gonzales,, MS, CFY-SLP State University of New York New Paltz INTRODUCTION Time is not a thing that, like an apple, may be perceived. Stimuli and patterns of stimuli occupy physical time; and we react to such stimuli by perceptions, judgments, comparisons, etc.... it follows that time is a concept, somewhat like the value of pieces of money, that attaches to perceptions only through a judgmental process. Woodrow, 1951 (from E.W. Large) 2 INTRODUCTION Perception of rhythm is a complex phenomenon related to the perception of time Relatively little is known about the perception of the rhythmic basis of connected speech and its developmental course over the lifespan (see Corriveau, Pasquini, & Goswami, 2007), although extensive work has been done on the acquisition of prosody as a whole of which rhythm is one component (Gerken, & McGregor, 1998; Kehoe, & Stoel-Gammon, 1997; Wells, Peppe, & Goulandris, 2004) 3 1
background Rhythm pervades experience from breathing to walking, to swimming and swallowing indicating a strongly physically constrained aspect Rhythm is part of music and language Pervasive rhythm patterns of both music and language vary culturally, indicating a likely learned aspect 4 background Western rhythm tends to be isochronous: characterized by an underlying meter of... multiple hierarchical levels of evenly spaced periodic structure, with faster levels resulting from binary or tertiary subdivisions (Hannon & Trehub, 2005) Non-Western rhythm may be nonisochronous, i.e., consist of... alternating short and long temporal intervals 5 background & PURPOSE P Here we examine detection of rhythm disruptions in isochronous vs non-isochronous pieces Ultimately looking at the relationship of rhythm and language in adults Recognition of disruptions in rhythmic patterns requires robust representations held in working memory for comparison to variant(s) or online recognition of pattern? 6 2
background Hannon, E. E. & Trehub, S.E. (2005): 6-month old infants compared sequences of familiar and non-familiar rhythms and detected disruptions in these sequences without special training 12-month old infants did so only with direct exposure Adult accuracy measured at chance levels The current study examined the plasticity of rhythm disruption detection by looking at a sub-population of adults with special training 7 METHOD ETHOD Participants & Stimuli Participants - 13 adult volunteers, all with some background in the study of speech - Age ranges from 20-66 years (M=37.38 yrs, median = 37) - Most reported no special training in music Participants were exposed to two rhythmic sequences 8 METHOD ETHOD... Stimuli Rhythmic patterns typical of isochronous (Western) and nonisochronous non-western music Two base rhythm patterns were played on a MIDI keyboard and recorded using Sony Acid Pro sequencing software. The MIDI data triggered the Afro-Cuban Kit setting of the Native Instruments Battery 2 percussion software plug-in. The first pattern was an eight- measure sequence with four beats per measure. Lower pitched percussion played a simple pattern with a heavy pulse on beats one and three. Higher pitched percussion played a steady eighth note pulse. This is a common sound in Western music. 9 3
METHOD ETHOD... Stimuli The second pattern was also an eight-measure sequence, however it contained five beats per measure. Lower pitched percussion played a pattern with a heavy pulse on beats one, three, and five. Higher pitched percussion kept an eighth note pulse, but the pattern was altered to accommodate the extra beat. This pattern is indicative of non- Western music. 10 METHOD Procedure Participants were then asked to listen to the same sequences which had been manipulated by inserting or deleting a beat Participants were asked to (non-verbally) indicate by pointing whether (same/different) and how (addition/deletion) the new sequence differed from the original 11 RESULTS Findings indicate stronger than predicted performance in rhythm disruption detection stimulus/analysis mean trials (n/6) correct isochronous 4.3846 nonisochronous 4.7692 t-distribution, df = 12 0.0004 0.0002 12 4
RESULTS Better than expected performance on nonfamiliar (nonisochronous) patterns t-test: Paired Two Sample for Means, df=12: mean Variance Pearson Correlation Hypothesized Mean Difference t Stat iso 4.3846 2.0897 0.1333 0-0.7495 noniso 4.7692 1.8590 P(T<=t) onetail 0.2340 t Critical onetail 1.7823 13 DISCUSSION Compare with Hannon & Trehub adult experiment: Different stimuli: H&T used music with both melody and beat ( Mary Had a Little Lamb for the isochronous and Balkan music for the nonisochronous): easier, harder, or confounding? Scoring: H&T used a 5 point similarity judgment scale Participants were paid volunteers selected from the college student population Included a training session (as with children) Instructed to listen to a CD at home twice daily (by selfreport most listened once daily) 14 DISCUSSION...H&T.. H&T Results: adults scored better in the isochronous condition than in the nonisochronous training made little difference, no correlation between amount of training and accuracy adult accuracy remained at chance levels for both conditions 15 5
Implication.. Direction A rhythm disruption detection task appears to be worthwhile research task with possible applications in assessment and treatment Plasticity/learnability? Different propensities? Investigate effect of training vs no training Compare with different but related skills 16 Special Thanks Thanks to Mike Gonzales for all his help preparing and describing the rhythm detection stimuli Thanks to Sarah Tempestilli for her assistance in organizing this presentation Special thanks to all of the S.U.N.Y students, faculty, and staff who enthusiastically supported this venture For references or more info contact: hestere@newpaltz.edu 17 6