The aim of this study was to develop a paradigm for obtaining a multi-feature profile for central auditory processing of different magnitudes of prosodic and phonetic changes in speech sounds. We recorded the MMNs to three vowel identity changes, three magnitudes of changes in intensity, and vowel duration as well as to two magnitudes of pitch changes from semi-synthetic vowels in 34min. Furthermore, we examined how the type and magnitude of deviation affect the size and timing of the MMN. All sound changes elicited statistically significant MMN responses, with the MMNamplitudes increasing with an increase in sound deviance. Importantly, the MMN amplitudes for the vowel changes reflected the differences between the phonemes, as did the MMNs to vowel-duration changes reflect the categorization of these sounds to short and long vowel categories, which are meaningful in the Finnish language. This new multi-feature MMN paradigm is suitable for investigating the central auditory processing of different magnitudes of speech-sound changes and can be used, for instance, in the investigation of pre-attentive phoneme categorization. The paradigm is especially useful for studying speech and language disorders in general, language development, and evolution of phoneme categories early in life, as well as brain plasticity during native or second language learning
Ambiguity plays an important role in our everyday cognitive experience. Since the 1980s, the neural bases for the perception of ambiguous information have been investigated but remains poorly understood. In our previous research, an increase of the N400 ERP component was found to be a common response for the perception of two different types of ambiguous stimuli: “canned” verbal jokes and ambiguous figures (Shcherbakova, Filippova, 2016; Filippova, Shcherbakova, Shtyrov, 2018). The current experiment aimed to understand the relationship between the error related negativity (ERN) component arising from jokes and ambiguous figures mistaken for non-humoristic texts and non-ambiguous figures.
Fourteen participants (9 females) went through two similar experimental procedures with 36 ambiguous and 36 non-ambiguous figures; 14 verbal jokes and 14 similar but non-humoristic short stories. Firstly, participants were presented with figures of both types and asked to identify whether each figure was ambiguous or non-ambiguous. We recorded ERPs that were timelocked to each answer about ambiguity/non-ambiguity of the figure presented. Secondly, participants were presented with the verbal stories and asked to identify whether each story was a joke or not. In this case, ERPs were time-locked to each answer about the key phrase of a joke/non-joke presented word-by-word on the computer screen after the whole text.
But we found an increase of the ERPs’ negativity in ambiguous figures that were mistaken for non-ambiguous ones in the ERN time window (Fz (F(3,622) = 12,6; p b 0.00) and Cz (F(3,625) = 6,96; p b 0.00)). Also, the results revealed no increase of the ERPs’ negativity in verbal jokes that were mistaken for non-jokes in the ERN time window. The results show that participants appeared to be sensitive (without awareness) to ambiguous figures that were identified as non-ambiguous ones. The level of this unconscious sensitivity is therefore reflected by the increases in negativity.
When a participant cannot correctly identify ambiguous stimulus at a conscious level, increases in negativity may be indexing greater violations of incongruence within an internal representation of meaning. These violations may precede semantic reversion of ambiguous figures and the understanding of a joke’s meaning.
Supported by RFBR (Dpt of Humanities and Social Sciences) grant #17-06-01014 А and RFBR grant #18-013-01086.