Assessment of auditory statistical learning by magnetic frequency tagged responses
-
1
Université Libre de Bruxelles, Belgium
-
2
Université Libre de Bruxelles, Belgium
-
3
Université Libre de Bruxelles, Belgium
Introduction: Auditory statistical learning (SL) is the incidental and automatic learning of transition probabilities embedded in a continuous stream of stimuli into subparts (usually triplets of stimuli) always presented together [1]. Standard behavioral test intended to assess the learning can be criticized, and implicit as well as on-line measures might be of better interest to assess learning [2]. Changes in electromagnetic frequencies are of great interest to assess SL and might represent a faster technique, less sensitive to the baseline type, artifacts and/or paradigm compared to evoked related potentials [3]. Based on the hypothesis that the brain electromagnetic activity responds to stimulation by resonating at the frequency of stimulation i.e. steady state potentials, we hypothesize that after learning, the resonance frequency during presentation of subparts-based stream of stimuli will be the one of the triplets and no longer of the tones. In other words, participants will no longer process the information as a stream of tones, but rather as a stream of tritones. Using magnetoencephalography (MEG), we aimed at validating the steady state approach to assess learning and isolate the neural bases of statistical learning and to reconstruct neural sources for this switch of resonance frequency power.
Methods: Twelve pure sounds (150 ms duration) within the same octave were randomly assigned to either 1) a statistical (STAT) stream or 2) a random (RDM) stream. Transitional probabilities (TPs) between tones were the unique differences between STAT and RDM. In STAT, tones were grouped into 4 tritones (intra-tritone TPs: 100%; inter-tritone TPs: 33%) while in RDM, tones were randomly presented (TPs: 9%). For both streams, tones were separated with a 25 ms pause. Moreover, a “subliminal pause” of 20 ms was added every three tones (i.e. between tritones in STAT) to enhance segmentation. Tones were therefore delivered at a 5.5 Hz frequency while tritones were delivered at a 1.8 Hz frequency. To increase the implicit character of the learning, STAT and RDM streams were intermixed and delivered continuously (STAT-RDM-STAT-RDM, counterbalanced). In total, each sound was repeated 280 times and the exposure phase lasted 20 min (4x5min). Five right-handed and healthy participants (2 females, mean age: 25.6 years) were tested in a MEG setting (whole-scalp MEG, Elekta). They were instructed to listen carefully to the sounds without any instruction about learning. Continuous MEG data were averaged over the two streams for each condition (2x5 minutes) to decrease signal to noise ratio while keeping a high frequency resolution (3.3mHz). Spectral analysis (1-12 Hz) of both STAT and RDM streams were performed using wavelet transform (Fieldtrip).
Results: A peak in the power spectrum was clearly apparent at the tone frequency (5.5 Hz) in both conditions. Moreover, a peak in the tritone frequency (1.8 Hz) was only present in the STAT condition. The difference in power spectrum between STAT and RDM (averaged across all subjects) showed a clear increase at 1.8 Hz and its harmonics, together with a clear decrease at 5.5 Hz in bilateral temporal area using combined gradiometers (see fig. 1). These differences were observed in 4 out of the 5 participants. This shift of resonance frequency was particularly prominent in right temporal channels, although it was observed bilaterally. Participants were not aware of the presence of tritones in streams and performed at chance at a recognition test.
Conclusion: These results confirm our hypothesis that participants no longer process repetitions of tones, but rather repetitions of tritones after exposure to statistical regularities. Moreover, steady-state analyses prove robust to assess statistical learning, and might provide an individual neural marker of implicit learning.
Acknowledgements
This work is supported by a grant from the Belgian National Fund for Scientific Research (FNRS).
References
[1] J. R. Saffran, et al. “Statistical learning of tone sequences by human infants and adults,” Cognition, vol. 70, no. 1, pp. 27–52, Feb. 1999.
[2] C. François, B. et al. “Cognitive and methodological considerations on the effects of musical expertise on speech segmentation,” Ann. N. Y. Acad. Sci., vol. 1252, pp. 108–115, Apr. 2012.
[3] M. Buiatti, M. et al. “Investigating the neural correlates of continuous speech computation with frequency-tagged neuroelectric responses,” NeuroImage, vol. 44, no. 2, pp. 509–519, Jan. 2009.
Keywords:
statistical learning,
Magnetoencephalography,
steady-state,
frequency-tagged response,
implicit learning,
Incident learning
Conference:
Belgian Brain Council 2014
MODULATING THE BRAIN: FACTS, FICTION, FUTURE, Ghent, Belgium, 4 Oct - 4 Oct, 2014.
Presentation Type:
Poster Presentation
Topic:
Basic Neuroscience
Citation:
Farthouat
J,
Op De Beeck
M,
Mary
A,
Delpouve
J,
Leproult
R,
Franco
A,
De Tiège
X and
Peigneux
P
(2014). Assessment of auditory statistical learning by magnetic frequency tagged responses.
Conference Abstract:
Belgian Brain Council 2014
MODULATING THE BRAIN: FACTS, FICTION, FUTURE.
doi: 10.3389/conf.fnhum.2014.214.00047
Copyright:
The abstracts in this collection have not been subject to any Frontiers peer review or checks, and are not endorsed by Frontiers.
They are made available through the Frontiers publishing platform as a service to conference organizers and presenters.
The copyright in the individual abstracts is owned by the author of each abstract or his/her employer unless otherwise stated.
Each abstract, as well as the collection of abstracts, are published under a Creative Commons CC-BY 4.0 (attribution) licence (https://creativecommons.org/licenses/by/4.0/) and may thus be reproduced, translated, adapted and be the subject of derivative works provided the authors and Frontiers are attributed.
For Frontiers’ terms and conditions please see https://www.frontiersin.org/legal/terms-and-conditions.
Received:
01 Jul 2014;
Published Online:
13 Jul 2014.
*
Correspondence:
Miss. Juliane Farthouat, Université Libre de Bruxelles, Bruxelles, Belgium, juliane.farthouat@ulb.ac.be