| 1. |
- Shirnin, Denis, 1983-, et al.
(författare)
-
Speech perception in noise : prediction patterns of neural pre-activation in lexical processing
- 2017
-
Konferensbidrag (övrigt vetenskapligt/konstnärligt)abstract
- The purpose of this study is to examine whether the neural correlates of lexical expectations could be used to predict speech in noise perception. We analyse mag-netoencephalography (MEG) data from 20 normal hearing participants, who read a set of couplets (a pair of phrases with rhyming end words) prior to the experiment. During the experiment, the participants are asked to listen to the couplets, whose intelligibility is set to 80%. However, the last word is pronounced with a delay of 1600 ms (i.e. expectation gap) and is masked at 50% of intelligibility. At the end of each couplet, the participants are asked to indicate if the last word was cor-rect, i.e. corresponding to the expected word. Given the oscillatory characteristics of neural patterns of lexical expectations during the expectation gap, can we predict the participant’s actual perception of the last word? In order to approach this re-search question, we aim to identify the correlation patterns between the instances of neural pre-activation, occurring during the interval of the expectation gap and the type of the given answer. According to the sequential design of the experiment, the expectation gap is placed 4400 ms prior to the time interval dedicated to the participant’s answer. Machine Learning approach has been chosen as the main tool for the pattern recognition.
|
|
| 2. |
|
|
| 3. |
- Signoret, Carine, 1981-, et al.
(författare)
-
Resolving discrepancies between incoming auditory information and linguistic expectations
- 2018
-
Ingår i: Neuroscience 2018. - : Society for Neuroscience.
-
Konferensbidrag (övrigt vetenskapligt/konstnärligt)abstract
- Speech perception in noise is dependent on stimulus-driven and knowledge-driven processes. Here we investigate the neural correlates and time course of discrepancies between incoming auditory information (i.e. stimulus-driven processing) and linguistic expectations (knowledge-driven processing) by including 20 normal hearing adults in a MEG study. Participants read 48 rhyming sentence pairs beforehand. In the scanner, they listened to sentences that corresponded exactly to the read sentences except that the last word (presented after 1600 millisecond delay and with 50% intelligibility) was only correct in half of the cases. Otherwise, it was 1) phonologically but not semantically related, 2) semantically but not phonologically related, or 3) neither phonologically nor semantically related to the sentence. Participants indicated by button press whether the last word matched the sentence they had read outside the scanner. Behavioural results showed more errors in condition 1 than in conditions 2 or 3, suggesting that phonological compatibility overrides semantic discrepancy when intelligibility is poor. Event-related field analysis demonstrated larger activity on frontal sites for correct than unrelated words, suggesting that the former were more accurately expected than the latter. An early M170 component was also observed, possibly reflecting expectation violation in the auditory modality. Dipole analysis will reveal whether M170 could be modulated by type of linguistic discrepancy. Distributed-network analysis will further our understanding of the time course and neural correlates of discrepancies between incoming auditory information and linguistic expectations.
|
|
| 4. |
- Stenfelt, Stefan, 1969-, et al.
(författare)
-
Auditory, signal processing, and cognitive factors influencing speech perception in persons with hearing loss fitted with hearing aids – the N200 study
- 2016
-
Konferensbidrag (övrigt vetenskapligt/konstnärligt)abstract
- Objective: The aim of the current study was to assess aided speech-in-noise outcomes and relate those measures to auditory sensitivity and processing, different types of cognitive processing abilities, and signal processing in hearing aids.Material and method: Participants were 200 hearing-aid wearers, with a mean age of 60.8 years, 43% females, with average hearing thresholds in the better ear of 37.4 dB HL. Tests of auditory functions were hearing thresholds, DPOAEs, tests of fine structure processing, IHC dead regions, spectro-temporal modulation, and speech recognition in quiet (PB words). Tests of cognitive processing function were tests of phonological skills, working memory, executive functions and inference making abilities, and general cognitive tests (e.g., tests of cognitive decline and IQ). The outcome test variables were the Hagerman sentences with 50 and 80% speech recognition levels, using two different noises (stationary speech weighted noise and 4-talker babble), and three types of signal processing (linear gain, fast acting compression, and linear gain plus a non-ideal binary mask). Another sentence test included typical and atypical sentences with contextual cues that were tested both audio-visually and in an auditory mode only. Moreover, HINT and SSQ were administrated.Analysis: Factor analyses were performed separate for the auditory, cognitive, and outcome tests.Results: The auditory tests resulted in two factors labeled SENSITIVITY and TEMPORAL FINE STRUCTURE, the cognitive tests in one factor (COGNITION), and the outcome tests in the two factors termed NO CONTEXT and CONTEXT that relates to the level of context in the different outcome tests. When age was partialled out, COGNITION was moderately correlated with the TEMPORAL FINE STRUCTURE and NO CONTEXT factors but only weakly correlated with the CONTEXT factor. SENSITIVITY correlated weakly with TEMPORAL FINE STRUCTURE and CONTEXT, and moderately with NO CONTEXT, while TEMPORAL FINE STRUCTURE showed weak correlation with CONTEXT and moderate correlation with NO CONTEXT. CONTEXT and NO CONTEXT had a moderate correlation. Moreover, the overall results of the Hagerman sentences showed 0.9 dB worse SNR with fast acting compression compared with linear gain and 5.5 dB better SNR with linear gain and noise reduction compared with only linear gain.Conclusions: For hearing aid wearers, the ability to recognize speech in noise is associated with both sensory and cognitive processing abilities when the speech materials have low internal context. These associations are less prominent when the speech material has contextual cues.
|
|
