You are here

Warning message

Attention! This event has already passed.

Cognitive processes in Second Language Acquisition: The role of noticing, attention and awareness in processing words in written L2 input

Tuesday, 14 December, 2010 - 14:00
Campus: Brussels Humanities, Sciences & Engineering campus
Faculty: Arts and Philosophy
Aline Godfroid
phd defence

This Ph.D. dissertation focuses on the construct of ‘noticing’ in Second
Language Acquisition Research (SLAR) and on ways of capturing ‘noticing
events’ that involve unfamiliar words in written second-language (L2) texts.
A key concept in the Second Language Acquisition (SLA) literature, noticing
denotes the cognitive process of paying attention to and becoming aware of
a (typically new) language form in the input, such as an unfamiliar word or
an unknown irregular verb form (p. 3). The presumably crucial role of
noticing for SLA is at the heart of Schmidt’s Noticing Hypothesis, which states
that noticing is the “necessary and sufficient condition for the conversion of
input to intake” (Schmidt, 1990: 129). Furthermore, given the hybrid nature
of noticing as a process that combines (a certain quality of) attention and (a
certain level) of awareness, we argue that empirical studies of noticing in
SLA must be clear in whether primary evidence is sought at the level of
attention (‘noticing as attention’) or at the level of awareness (‘noticing as
awareness’; chapters 1 and 3).

Prior to this dissertation, the role of noticing in SLA had been investigated
almost exclusively with respect to the construct of noticing as awareness,
with much less attention to noticing as attention (chapter 3). While previous
research has been productive and illuminating, each method has its
drawbacks. In particular, verbal reports have been criticised for both nonveridicality
(when gathered offline) and reactivity (when gathered online).
By contrast, eye-tracking data — justified by the eye-mind assumption (the
assumption that overt attention and covert attention are tightly linked;
Reichle, Pollatsek & Rayner, 2006) — have not been brought to bear on
these problems until now (chapter 3).

After the potential of eye tracking as an instrument for measuring noticing
was assessed and confirmed in a pilot study (n = 9; chapter 4), a different
sample of 28 advanced learners of English (L1: Dutch) took part in the main
study (chapter 5). These participants read 20 paragraphs in English, 12 of
which contained a target area in one of the following four conditions: (I)
known, existing word (e.g. boundaries), (II) pseudo word (e.g. paniplines),
(III) pseudo word followed by a semantically constraining existing word (e.g.
paniplines or boundaries) and (IV) pseudo word preceded by a semantically
constraining existing word (e.g. boundaries or paniplines).

Analysis of the eye-tracking data revealed evidence for increased attention to
the new word forms, even though task instructions had encouraged
participants to focus on text meaning. Specifically, we found that:

1. learners spent extra time processing the unknown pseudo words
(experimental conditions II, III and IV) in comparison to the matched
controls (control condition I);
2. learners invested even more time trying to process the unknown word
when immediately afterwards it was followed by a known synonym
(condition III). This effect was reflected in longer processing times for
the known word rather than in further increases in processing time for
the novel word.
3. the longer learners fixated on the unknown word (all experimental
conditions), the more likely they were to recognise that word on an
unannounced, immediate vocabulary post-test;
4. the extra processing time observed for the clarifying known word in
condition III did not yield better recognition of the preceding unknown
word on the vocabulary post-test.

In the concluding discussion of the dissertation (chapter 6), we point towards
an integration of Robinson’s (1995, 2003) model of noticing in SLA with
Dehaene and associates’ Global Neuronal Workspace model (e.g. Dehaene &
Changeux, 2004) and the classic Craik and Lockhart (1972) levels-ofprocessing
framework. A central claim is that the concept of noticing should
be viewed, not as an all-or-nothing event, but as a multidimensional process
defined by a sequence of events that involve attention and awareness. The
quantity (duration) and quality (nature of mental operations) of the noticing
process, then, are hypothesised as predictors of memory strength, i.e. as
predictors of the amount of long-term learning.