Impaired emotion recognition in music in Parkinson’s disease

It is well known that music has the power to evoke strong emotions and, perhaps not surprisingly, music is often used in psychological studies for the purposes of examining the effects of emotional moods or states on behaviour. Another use for music that is becoming more common is testing ability to recognise emotions. The latter premise was used as the basis for a new study out this month in Brain and Cognition.

The background: Studies have shown that patients with Pakinson’s disease (PD) have impairments in emotion recognition, but so far this deficit had only been tested with faces and voices (emotional prosody). The results have not always been consistent in showing which emotions are affected, neither have all studies demonstrated whether emotion recognition difficulties in PD relate to other cognitive impairments, such as difficulties with executive function (attention, decision making etc).  Executive function has shown to be important in the recognition of emotion (Grey & Tickle-Degnen, 2010) and PD patients frequently show executive dysfunctions (Dubois & Pillon 1997). So it is important to work out how much the emotional recognition deficit in PD is a unique effect.

The hypotheses:

1)      What would happen if patients were required to recognise emotion in music? The authors hypothesised that emotion recognition would be impaired, as it is with other stimuli.

2)      They hypothesised that the emotion recognition effects would be stronger with music compared to those previously seen with faces and voices, as music is often associated with stronger emotions that faces or voices (Goldstein, 1980).

3)      Lastly the authors carried out a number of other tests designed to measure executive function in their PD patients. They wanted to know if problems with emotion recognition for music could be due to wider difficulties with executive function, or whether musical emotion recognition problems were a unique non-cognitive effect of PD.

The tasks: For the emotion recognition part of the experiment participants listened to 32 musical experts divided up into 8 happy, 8 sad, 8 angry and 8 fear. The excerpts were selected from previous studies on the basis that their emotional expression been correctly recognised by control healthy subjects on at least 90% of occasions. The pieces were selected from Baroque, Classical and Romantic periods, non vocal, and lasted between 9-35 seconds (average = 17s). Participants simply listened to the excerpts over loud speakers and stated which of the four emotions they expressed. For the other cognitive function tests the authors tested first musical ability using two of the subtests from the MBEA (I think the scale and rhythm tests from my reading). Then, for wider executive function, they used Category fluency, Controlled Oral Word Association test, Trail Making test, Digit span (WAIS) and GIT visuospatial reasoning (refs available in the paper).

The findings:

1) MANOVA tests revealed that PD patients performed worse than controls on the emotion task, but that this effect was limited to the fear and anger music conditions. Both PD patients and controls were at ceiling for the happy and sad excerpts.

2) The effect size obtained by the music test in the present experiment was greater than that obtained in previous studies using faces and voices (Cohen’s d = 1.3 vs. 0.8 respectively)

3) Finally the analysis showed that the effects remained after covarying for executive function – meaning that some unique variance in the emotional recognition problems in the PD patients could not be explained by their lower scores on the executive function tests. Also, score on the fear tests was associated with executive function tests, implying that judging fear in music and executive function do have something in common.

Conclusions: This paper contributes to the literature on emotion recognition in PD by demonstrating the deficit in a new, potentially more powerful modality. The fact that the effect only occurs in negative complex emotions is supported by research in other modalities. The link between fear recognition and executive function is an interesting one which might have future implications for studying the brain structures that could be jointly active in these two tasks (i.e. amygdala).  Finally, whilst this study underlies the importance of studying executive function in PD, the finding of unique variance in emotion recognition difficulties implicates PD patients encounter problems in another, purely emotional part of the recognition process.

References:

Mirjam J. van Tricht, Harriet M.M. Smeding, Johannes D. Speelman and Ben A. Schmand (2010). Impaired emotion recognition in music in Parkinson’s disease. Brain and Cognition, 74(1), 58-65. doi:10.1016/j.bandc.2010.06.005

Dubois, B., & Pillon, B. (1997). Cognitive deficits in Parkinson’s disease. Journal of Neurology, 244, 2-8.

Goldstein, A. (1980). Thrills in response to music and other stimuli. Physiological Psychology, 8, 126–129.

Gray HM, & Tickle-Degnen L. (2010) A meta-analysis of performance on emotion recognition tasks in Parkinson’s disease. Neuropsychology, 24(2), 176-91. PMID: 20230112