Planning and Focusonform in TaskBased Language Learning

1
Planning and Focus-on-form in Task-Based Language
Learning

• Ryo Nitta
• University of Warwick
• TBLT Conference
• Leuven, 2005

2
Introduction

• Two research traditions in SLA planning and
  focus-on-form
• The study analysis of task performance and
  verbal protocols
• Findings of task performance analysis
• Implications
• Issues for the second stage

3
Types of Planning (Ellis, 2005)
Rehearsal (task repetition)
Pre-task planning
Strategic planning
Planning
Pressured
Within-task planning
Unpressured (on-line planning)
4
Strategic vs. On-line Planning

• Strategic planning
• Information-processing model
• Skill-learning model
• On-line planning
• L2 Monitoring (Morrison Law, 1983)
• Careful within-task planning (Yuan Ellis, 2003)

5
Summary of the Previous Research
6
Focus-on-Form Instruction

• The noticing issue Do learners have the
  cognitive resources to notice the gap between
  their IL utterances and the TL utterances around
  them?
• The interruption issue Is a pedagogical
  intervention that does not interrupt the
  learners own processing for language learning
  even possible?
• The timing issue If so, then precisely when,
  in cognitive terms, should the pedagogical
  intervention occur? (Doughty, 2001)

7
Identifying a Link between Planning and
Focus-on-Form

• Theoretical issue Strategic planning facilitates
  a learner-driven focus-on-form (Ortega, 1999,
  2005)
• Methodological issue A process-product approach
  to planning. How do L2 learners plan during
  strategic planning time? (Ortega, 1999, 2005
  Sangarun, 2005)

8
Research Questions

• Do different planning conditions differently
  influence the oral performance?
• How do L2 speakers plan their speech on-line when
  their performance is underway? Do they focus on
  form in on-line planning more frequently than
  strategic and no-planning conditions?

9
Research Hypotheses for Task Performance Analysis
(1)

• Hypothesis One Strategic planning (SP) will give
  more positive influence on fluency, complexity
  and accuracy than no-planning (NP).
• Hypothesis Two On-line planning (OP) will give
  more positive influence on complexity and
  accuracy, but more negative influence on fluency
  than NP.
• Hypothesis Three OP will give more positive
  influence on complexity and accuracy, but more
  negative influence on fluency than SP.

10
Research Hypotheses for Task Performance Analysis
(2)

• Hypothesis Four The effects of SP will interact
  with the different proficiency levels.
• Hypothesis Five The effects of OP will interact
  with the different proficiency levels.

11
Research Design
12
Task

• Story-telling tasks x 3 6 picture cartoons taken
  from a popular story-telling resource book for
  EFL learners (Heaton, 1975).
• Different, but clearly structured tasks (cf.
  Tavakoli Skehan, 2005)
• The first obligatory sentence was given to
  encourage the use of past tense forms in each
  task, following Ellis (1987).

13
Task Implementation
14
Participants

• 27 Japanese speakers of English (male 11,
  female 16)
• Length of residence mean 11.86 (months), SD
  13.74, Range 0.25 (i.e., 1 week) - 41.00 (i.e.,
  3 years and 5 months)
• High vs. Low Proficiency 6-point global ratings
  by three raters to judge the NP speech. The
  average scores were used to categorize high and
  low proficient groups.

15
Fluency

• Pruned speech rates
• Mean length of run
• Total length of pauses per minute
• Number of end-clause pauses per 100 words
• Number of mid-clause pauses per 100 words
• Number of filled pauses
• Number of dysfluencies (i.e., false-starts,
  reformulation, repetition and self-correction)

16
Complexity

• Syntactic complexity the number of clauses per
  AS-unit (Foster, Tonkyn and Wigglesworth, 2000)
• Discoursal complexity the number of discourse
  organization devices (Ejzenberg, 2000)
• Chaining integration devices
• Grammatical integration devices

17
Accuracy

• Global measures percentage of error-free clauses
  (e.g., Foster Skehan, 1996)
• Specific measures
• Percentage of target-like verb forms (Yuan
  Ellis, 2003)
• Percentage of target-like article (Crookes, 1989)

18
Factor Analysis

• On the whole, the results of three factor
  analyses represent three independent dimensions,
  fluency, complexity and accuracy.
• Mean length of run was selected for MANOVA
  analysis as a fluency component.
• The number of grammatical integration devices was
  selected as a complexity component.
• The percentage of error-free clauses was selected
  as an accuracy component.

19
Repeated measures MANOVA
20
Univariate test of within-participant effect
21
ANOVAs in Two Proficiency Groups
22
Specific Accuracy Measures
23
Implications

• Trading-off between fluency and form
• Fluency and accuracy
• Fluency and complexity
• Interaction between fluency and proficiency
• Monologic story-telling task type?
• Dual mode?
• Interaction between accuracy and proficiency
• Form-focused effects of on-line planning

24
Next Stage

• To identify the mental operations in the three
  planning conditions.
• To identify different mental operations between
  high and low proficient speakers.
• Strategic planning ? High proficient speakers
• On-line planning ? Low proficient speakers