Elsevier

Brain and Cognition

Volume 119, December 2017, Pages 45-53
Brain and Cognition

Jazz musicians reveal role of expectancy in human creativity

https://doi.org/10.1016/j.bandc.2017.09.008Get rights and content

Highlights

  • Jazz musicians prefer unexpected chord progressions.

  • Classical and Jazz musicians show different brain responses to unexpected events.

  • Event Related Potentials to unexpected events correlate with creative behavior.

Abstract

Creativity has been defined as the ability to produce work that is novel, high in quality, and appropriate to an audience. While the nature of the creative process is under debate, many believe that creativity relies on real-time combinations of known neural and cognitive processes. One useful model of creativity comes from musical improvisation, such as in jazz, in which musicians spontaneously create novel sound sequences. Here we use jazz musicians to test the hypothesis that individuals with training in musical improvisation, which entails creative generation of musical ideas, might process expectancy differently. We compare jazz improvisers, non-improvising musicians, and non-musicians in the domain-general task of divergent thinking, as well as the musical task of preference ratings for chord progressions that vary in expectation while EEGs were recorded. Behavioral results showed for the first time that jazz musicians preferred unexpected chord progressions. ERP results showed that unexpected stimuli elicited larger early and mid-latency ERP responses (ERAN and P3b), followed by smaller long-latency responses (Late Positivity Potential) in jazz musicians. The amplitudes of these ERP components were significantly correlated with behavioral measures of fluency and originality on the divergent thinking task. Together, results highlight the role of expectancy in creativity.

Introduction

One of the most striking features of the human brain is its ability to be creative. Creativity has been defined as the ability to produce work that is novel, high in quality, and appropriate to an audience (Sternberg, Lubart, Kaufman, & Pretz, 2005). While the nature of the creative process is under debate, many believe that creativity relies on real-time combinations of known mental processes (Goldenberg, Mazursky, & Solomon, 1999), with contributions from the society and culture as well as from the person (Csikszentmihalyi, 1996). However, how these neural and cognitive processes are combined is unknown, as they vary across domains and between individuals.

One model of creativity in real time comes from musical improvisation, such as in jazz music, in which individuals spontaneously create novel auditory-motor sequences that are aesthetically and emotionally rewarding (Bengtsson et al., 2007, Berkowitz and Ansari, 2008, Limb and Braun, 2008, Liu et al., 2012). Jazz improvisers show higher divergent thinking ability and openness to experience, even when compared to musicians with other types of training (Benedek, Borovnjak, Neubauer, & Kruse-Weber, 2014). Longitudinal studies have also shown that improvisation training induces improvements in performance on divergent thinking tasks (Karakelle, 2009, Lewis and Lovatt, 2013). Due to its reliance on domain-general as well as domain-specific processes, the study of improvisation is thought to have implications not only for the study of artistic expertise, but also for the neural underpinnings of domain-general processes such as motor control and language production (Beaty, 2015).

While the mechanisms of creativity are unclear, recent work from theoretical and modeling studies suggests that the processing of deviance, or of unexpected events, is key to creativity (Kleinmintz et al., 2014, Wiggins and Bhattacharya, 2014). If expectation processing is key to creativity, one would expect that individuals with more training in creativity might process expectancy differently. Here we aim specifically to inspect the role of expectation in creativity, using jazz improvisers as a model.

Across multiple modalities, expectation violations from novel stimuli elicit the P3, a positive ERP as measured using event-related potentials as a peak around 300–600 ms after the onset of target events (Arthur and Starr, 1984, Klein et al., 1984, Knight et al., 1989, Yamaguchi and Knight, 1991). The P3 is elicited across multiple sensory domains and has generally been linked to engagement, arousal, and novelty detection (Friedman, Cycowicz, & Gaeta, 2001; Murphy, Robertson, Balsters, & O'Connell R, 2011). It includes two subcomponents, P3a and P3b (Polich, 2007). P3a (or Novelty P3) is thought to reflect more stimulus-based attention and novelty detection in the frontal lobe, whereas P3b reflects attention- and memory-dependent neuroinhibitory processes especially in the parietal lobe (P3b) (Murphy et al., 2011, Polich, 2007). The P3 can be followed by an additional parietally-centered late positive potential (LPP), around 400–900 ms, which reflects evaluation and affective appraisal, especially for motivating and task-relevant events (Cacioppo et al., 1993, Schupp et al., 2000).

In addition to the P3 and the late positivity, expectation violations for musical harmony, which has been linked to emotion and meaning in music (Meyer, 1956), additionally elicits an Early Right Anterior Negativity (Koelsch et al., 2000, Loui et al., 2009, Steinbeis et al., 2006). This Early Right Anterior Negativity (ERAN) bears some similarities to the Mismatch Negativity (Naatanen, Simpson, & Loveless, 1982) in that both are sensitive to unexpected acoustic events and may reflect auditory prediction and comparison; however the ERAN is thought to be more specific to the processing of musical syntax and is sensitive to learning and experience (Koelsch, 2009, Loui et al., 2009).

Our general hypothesis is that creativity depends on sensitivity to unexpected events. Specifically, in the domain of music, we expect that jazz improvisers will process unexpected musical stimuli with increased sensitivity and engagement, as indexed by the ERAN and P3, compared to their musician and non-musician counterparts. Furthermore, we expect that neural indices of unexpectedness will be correlated with measures of divergent thinking.

Section snippets

Subjects

36 subjects (12 female) participated in the study. Subjects were recruited from Wesleyan University and the Hartt School of Music in exchange for compensation or course credit. Jazz improvising musicians, non-improvising (Classical) musicians, and non-musicians were recruited based on their reported musical experience. All three groups (n = 12 each, sample size determined from previous studies (Loui et al., 2005, Loui et al., 2009)) were matched in age, general intellectual function as assessed

Decreased preference for expected stimuli in Jazz musicians

Behavioral results from preference ratings for high expectation, medium expectation, and low expectation chord progressions showed a main effect of expectation (F(2,66) = 17, p < 0.001, η2 = 0.301), and an interaction between group and expectation (F(4,66) = 3.2, p = 0.018, η2 = 0.113) (Fig. 1B). The non-musicians strongly preferred the high expectation condition followed by the medium (t(23) = 2.3, p = 0.03, d = 0.618) and low expectation conditions (high vs. low: t(23) = 3.5, p = 0.0019, d = 1.841; medium vs. low:

Discussion

Using Jazz improvisation training as a model, we show for the first time that jazz improvising musicians have higher preference and markedly different neural sensitivity to unexpected musical stimuli. These neural measures of expectations in a musical context are correlated with measures of domain-general creativity. Based on these results, we posit that creative perception and cognition depends on sensitivity and engagement to unexpected events within the relevant domain.

Expectations are

Acknowledgements

Supported by grant funding from the Imagination Institute and the Grammy Foundation to PL. We thank Pheeroan Aklaff for helpful conversations, Sarah Knight and Tedra James for scoring responses to the TTCT, and all our participants.

References (40)

  • S. Yamaguchi et al.

    P300 generation by novel somatosensory stimuli

    Electroencephalography and Clinical Neurophysiology

    (1991)
  • D.L. Arthur et al.

    Task-relevant late positive component of the auditory event-related potential in monkeys resembles P300 in humans

    Science

    (1984)
  • A. Baddeley

    Working memory: Looking back and looking forward

    Nature Reviews Neuroscience

    (2003)
  • S.L. Bengtsson et al.

    Cortical regions involved in the generation of musical structures during improvisation in pianists

    Journal of Cognitive Neuroscience

    (2007)
  • D.E. Berlyne

    Aesthetics and psychobiology

    (1971)
  • M. Biasutti

    Pedagogical applications of the cognitive research on music improvisation

    Frontiers in Psychology

    (2015)
  • J.T. Cacioppo et al.

    If attitudes affect how stimuli are processed, should they not affect the event-related brain potential?

    Psychological Science

    (1993)
  • M. Csikszentmihalyi

    Creativity: Flow and the psychology of discovery and invention

    (1996)
  • A.D. Friederici

    The time course of syntactic activation during language processing: A model based on neuropsychological and neurophysiological data

    Brain and Language

    (1996)
  • J. Goldenberg et al.

    Creative sparks

    Science

    (1999)
  • Cited by (41)

    • Prediction under uncertainty: Dissociating sensory from cognitive expectations in highly uncertain musical contexts

      2021, Brain Research
      Citation Excerpt :

      As multiple studies have shown, the MMN to deviating sounds in an auditory sequence is significantly enhanced in musicians as compared with non-musicians (Tervaniemi et al., 2014; Vuust et al., 2011; Vuust et al., 2012) and even discriminates between different types of musicians (Vuust et al., 2012). Expertise effects on the P3b in response to unexpected stimuli have been observed (Przysinda et al., 2017), whereas a recent study did not find an interaction between expertise and entropy (Quiroga-Martinez et al., 2019). However, testing musicians with our more carefully controlled stimuli may reveal whether expertise-related interactions exist in a more ecological, high-uncertainty musical context.

    • Spatiospectral brain networks reflective of improvisational experience

      2021, NeuroImage
      Citation Excerpt :

      Presumably as a result of the specific nature of this training, a classically trained musician who may have trained playing an instrument just as many years as an improviser - just in a different way - may not be able to improvise music. Previous work found that, relative to classically trained musicians, jazz improvisers showed more pronounced and larger early right anterior negativity (ERAN) to rare and unexpected targets (Przysinda et al., 2017). Magnitudes of these ERAN responses correlated with metrics for improvisation experience and P3b and ERAN correlated with fluency and originality in divergent thinking tasks.

    View all citing articles on Scopus
    View full text