Caroline Lejeune, Xavier Schmitz, Stéphanie Lempereur, Christelle Maillart,
Caroline Lejeune, Xavier Schmitz, Stéphanie Lempereur, Christelle Maillart,
Thierry Meulemans & Audrey Gabriel
Thierry Meulemans & Audrey Gabriel
Thierry Meulemans, & Audrey Gabriel
Thierry Meulemans, & Audrey Gabriel
Department of Cognitive Science, University of
Department of Cognitive Science, University of Liège
Liège,
, Belgium
Belgium
It is generally admitted that implicit learning abilities are efficient early in childhood.
It is generally admitted that implicit learning abilities are efficient early in childhood. HOWEVER
HOWEVER, a question persists in the literature concerning the
, a question persists in the literature concerning the ageage--invarianceinvariance and the and the effects of sequence complexity effects of sequence complexity on learning. on learning.
OBJECTIVE
OBJECTIVE
OBJECTIVE
OBJECTIVE
ToTo exploreexplore, with two Serial Reaction Time tasks adapted to the young child, whether , with two Serial Reaction Time tasks adapted to the young child, whether implicit learning abilities implicit learning abilities are present to the same extent are present to the same extent in 4
in 4--, 7, 7--, and 10, and 10--yearyear--old childrenold children
To investigate the To investigate the impact of the structure impact of the structure of the sequence of the sequence onon the children’s performancethe children’s performance
To assess theTo assess the explicit awareness explicit awareness developed by the children in this 2 SRT tasksdeveloped by the children in this 2 SRT tasks
A TOTAL OF 107 CHILDREN
A TOTAL OF 107 CHILDREN ‐‐Monolingual French speakersMonolingual French speakers
I.I. ANOVA with Block (2 levels: 6 vs. 7) x Sequence (2 levels : ANOVA with Block (2 levels: 6 vs. 7) x Sequence (2 levels :
32 432 4--yearyear--old children old children --Normal or corrected visionNormal or corrected vision
39 739 7--yearyear--old old children children --No history of learning disabilitiesNo history of learning disabilities
36 1036 10--yearyear--old children old children --No neurological or behavioral diagnoses No neurological or behavioral diagnoses -- RightRight--handedhanded..
Serial Reaction Time (SRT): The «
Serial Reaction Time (SRT): The « Harry Potter taskHarry Potter task »» Participants were randomly assigned to one of two Participants were randomly assigned to one of two
SOC vs. Ambiguous) x Age group (3 levels: 4, 7, 10) SOC vs. Ambiguous) x Age group (3 levels: 4, 7, 10) Results showed
Results showed significant significant but but differentdifferent learninglearning effects for the two effects for the two types of sequences
types of sequences
Group effect: p<.001 Group effect: p<.001 ÎÎ 1010-- children are faster than 4children are faster than 4-- and 7and 7--yearyear--old childrenold children
Block effect : p<.001: block 6 is processed faster than the transfer block (B7)Block effect : p<.001: block 6 is processed faster than the transfer block (B7) Ambigous sequence 1600 2000 4 years 7 years 10 years SOC sequence 1200 1400 1600 4 years 7 years 10 years
Participants were randomly assigned to one of two Participants were randomly assigned to one of two
adapted SRT tasks: adapted SRT tasks:
Block x Sequence effect: p<.001 Block x Sequence effect: p<.001
Î
Î The dThe difference between block 6 and block 7ifference between block 6 and block 7 is greater for the
is greater for the Ambiguous sequenceAmbiguous sequence than for the SOC sequence
than for the SOC sequence
7 blocks (7 blocks (one block consisted of 8 repetitions of the sequence).one block consisted of 8 repetitions of the sequence).Î
Î the 7the 7ththblock = transfer block block = transfer block
The traditional SRT task was adapted in order to make the task more The traditional SRT task was adapted in order to make the task more attractive for the childrenattractive for the children
••
Ambiguous sequence Ambiguous sequence
(8 items) (8 items)
Second Order Conditional Second Order Conditional
sequence (12 items)
sequence (12 items) B1 B2 B3 B4 B5 B6 B7 Transfer
400 800 1200 Ms e c B1 B2 B3 B4 B5 B6 B7 Transfer 400 600 800 1000 Ms e c 800 850 900 950 1000 1050 1100 1150 sequence 8 sequence 12 II.
II. Learning Index measure (percent improvement) Learning Index measure (percent improvement)
Stadler (1995) ([Bloctransfert – Bloc 6]/ [Bloc transfert + Bloc 6])
SOC sequenceSOC sequence: No Age group effect: p = .42 : No Age group effect: p = .42 ÎÎ the lthe learning effect earning effect was similar in the different age groups
was similar in the different age groups
Ambiguous sequenceAmbiguous sequence: : Significant Age group effect : p=.047Significant Age group effect : p=.047 Î
Îthe the learning effect was different between age learning effect was different between age groups: the 10
groups: the 10--yearyear--old children showed better old children showed better learning effects than the 4
learning effects than the 4-- and 7and 7--yearyear--old children. old children. III
III Explicit awareness measuresExplicit awareness measures
attractive for the children attractive for the children
Î
Î
Touch screen Touch screen responding moderesponding modeÎ
Î
A sorcerer appears in four of the six windows of a castleA sorcerer appears in four of the six windows of a castleÎ
Î
The child has to touch the sorcerer with a magic wound (The child has to touch the sorcerer with a magic wound (pen styluspen stylus) )as fast and as accurately as possible as fast and as accurately as possible
Ambigous sequence 0,02 0,04 0,06 0,08 0,10 0,12 0,14 0,16 0,18 0,20 0,22 Lear ning I ndex BLOCK 6 BLOCK 7 750 III.
III. Explicit awareness measures Explicit awareness measures
Î
Îscore=score=the longest correct sequence produced by each child the longest correct sequence produced by each child
//
sequence’s lengthsequence’s length Group effect : p<.001: Group effect : p<.001: 1010--yearyear--oldold children are children are better than the other groups
better than the other groups
Sequence effect : p<.001: Sequence effect : p<.001: the generation score is the generation score is better for the
better for the ambiguous sequence ambiguous sequence than than for the SOC sequence
for the SOC sequence
No significant interaction: p =.153No significant interaction: p =.153
After the SRT taskAfter the SRT task, half of the children were asked a , half of the children were asked a series of questions series of questionsAND completed a
AND completed a free generate free generate task task to assess their subsequent explicit to assess their subsequent explicit awareness of the sequence (8 or 12 trials
awareness of the sequence (8 or 12 trials –– one repetition of the sequence)one repetition of the sequence)
4 years 7 years 10 years 0,00 Free generation sequence 8 sequence 12 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 4years 7years 10years 1)
1) 44--, 7, 7--, and 10, and 10--yearyear--old children demonstrate significant learning effects with both the old children demonstrate significant learning effects with both the ambiguousambiguous and the and the SOCSOC sequences with touch screen sequences with touch screen responding
responding ÎÎthese results confirm that implicit learning abilities are present in early childhoodthese results confirm that implicit learning abilities are present in early childhood 2)
2) An age effect An age effect appears only for the appears only for the ambiguous sequence ambiguous sequence ÎÎ the statistical structure of the sequence enables the child to use explicit the statistical structure of the sequence enables the child to use explicit strategies
strategies 3)
3) This latter result confirms that explicit strategies might intervene in the children’s SRT performance, which could explain tThis latter result confirms that explicit strategies might intervene in the children’s SRT performance, which could explain thehe age effect age effect observed for the ambiguous sequence
observed for the ambiguous sequence
C.Lejeune@ulg.ac.be