Information Technology Reference
In-Depth Information
located at the extreme positions on the continua were indeed perceived as typical
exemplars of their respective material categories, sounds in intermediate positions,
which were synthesized by interpolating the acoustic parameters characterizing
sounds at extreme positions, were consequently expected to be perceived as
ambiguous (e.g. to be neither wood nor metal). Participants were asked to cate-
gorize each of the randomly presented sounds as wood, metal or glass.
Based on the classi
cation rates, we de
ned
typical
sounds as sounds that
were classi
ed by more than 70 % of the participants in the right material category
and
ed by less than 70 % of the
participants in a given category. Ambiguous sounds were associated with slower
RTs than typical sounds. As might be expected, ambiguous sounds are therefore
more dif
ambiguous
sounds, those that were classi
cult to categorize than typical sounds. This result is in line with previous
findings in the literature showing that non-meaningful sounds were associated with
longer RTs than meaningful sounds. Electrophysiological data showed that
ambiguous sounds elicited more negative ERP (a negative component, N280,
followed by a negative slow wave, NSW) in fronto-central brain regions and less
positive ERP (P300 component) in parietal regions than typical sounds. This dif-
ference may re
culty to access information from long-term memory. In
addition, electrophysiological data showed that the processing of typical metal
sounds differed signi
ect the dif
cantly from those of typical glass and wood sounds as early
as 150 ms after the sound onset. The results of the acoustic and electrophysiological
analyses suggested that spectral complexity and sound duration are relevant cues
explaining this early differentiation. Lastly, it is worth noting that no signi
cant
differences were observed on the P100 and N100 components. These components
are known to be sensitive to sound onset and temporal envelope, re
ecting the fact
that the categorization process occurs in later sound-processing stages.
4.5.2 Conceptual Priming for Non-linguistic Sounds
In language, a comprehensible linguistic message is for instance conveyed by
associating words while respecting the rules of syntax and grammar. Can similar
links be generated between non-linguistic sounds so that any variation will change
the global information conveyed? From the cognitive neuroscience point of view,
one of the major issues that arises from this question is whether similar neural
networks are involved in the allocation of meaning in the case of language and that
of sounds of other kinds. In a seminal study using a priming procedure, Kutas and
Hillyard (Kutas and Hillyard 1980 ) established that the amplitude of a negative
ERP component, the N400 component, increases when
final sentence words are
incongruous (e.g. The
fish is swimming in the river/carpet ). Since then, the N400
has been widely used to study semantic processing in language. In recent studies,
priming procedures with non-linguistic stimuli such as pictures, odours, music and
environmental sounds have been used (e.g. Holcomb and McPherson 1994 ; Castle
et al. 2000 ; Koelsch et al. 2004 ; Daltrozzo and Sch
n 2009 ; Van Petten and
Rheinfelder 1995 ; Orgs et al. 2006 ). Although the results of these experiments
ö
 
Search WWH ::




Custom Search