Listening Cherry 06 – An impending revolution?

Listening Cherries is a blog where I talk about listening issues - from classroom activities to academic research. Listening Cherry no. 6 describes an impending revolution in speech perception research.

In my last blog here I reported feeling uneasy about the nature of research in speech perception, and in particular its use of a model/assumption known as ‘the phoneme input assumption’ (Magnuson et al., 2013).

But I feel  even more uneasy because I am open to the criticisms of this type: ‘You have no right to comment! You are cherry-picking the literature, and do not understand the whole picture.’

But interestingly, major figures in the field seem to be similarly – well not uneasy exactly – but aware that a change of approach (a paradigm shift in models) is required in their field. For example, Mirjam Ernestus (Ernestus, 2014) concludes that no model of speech perception is sufficiently developed ‘to obtain detailed insight in the human language capacity’ (p. 17).

Even more strongly, Magnuson et al (2013) argue that the field of research into speech perception is both on the verge of, and in need of, a  revolution. In their view, research is at a tipping point where the ‘temporary simplifying assumptions’ that the field has made in order to do research have now become obstacles – ‘complicating assumptions’ (Magnuson et al, 2013, p. 433).

They criticise certain aspects of the current models and state that researchers are:

constructing a pretend signal, rather than grappling with actual speech. Without tackling the signal, we will not know what helpful constraints we have hidden with the abstractions of our simplifying assumptions (p. 22).

The more I read this, the more I am horrified. ‘WHAT? You have been constructing a PRETEND SIGNAL for all these years? And you are NOT grappling with ACTUAL SPEECH????’

But I tell myself I should be grateful that this research is turning toward ‘actual speech’ and that it should (I hope) generate research findings which we can apply to language description, teacher training, and most of all into improving the teaching of listening.

Magnuson et al. go on to identify four ‘avenues as most promising for pushing the field beyond the tipping point and to new theoretical frameworks’ these are:

1. the need to grapple with the speech signal itself,
2. integration of the study of spoken word recognition with descriptively higher levels of language processing,
3. the need for theories and models to grapple with learning across the life span, including language development in childhood and rapid, flexible learning in adults, and
4. the need to respect neurobiological contraints on mechanisms for language processing. (p. 435/p. 23)

That looks to be a great agenda.

Ernestus, M. (2014). Acoustic reduction and the roles of abstractions and exemplars in speech processing. Lingua, 142, 27-41.

Magnuson, J. S., Mirman, D., & Myers, E. (2013). Spoken Word Recognition.The Oxford Handbook of Cognitive Psychology, 412.