A Learned Label Modulates Object Representations
in 10-Month-Old Infants
Katherine E. Twomey & Gert Westermann, Lancaster University, UK
• Infants begin to link their linguistic and nonlinguistic
representations in the first year, responding to familiar labels
from 6 months (Bergelson & Swingley, 2012)
• Converging evidence indicates labels encountered online
affect online nonlinguistic processing, e.g., guiding category
formation (Althaus & Westermann, 2016)
• Learned language also affects online category generalization
(e.g., Perry & Samuelson, 2011)
• In adults, labels shape cogntion (Lupyan, 2012), but do labels shape
nonlinguistic representations in infancy?
Gliga, Volein & Csibra, 2010 (E2)
• Trained 12-month-old infants with 2 novel 3D objects, one
with a novel label, one without
• Immediately afterwards, presented infants with images of (a)
labeled object, (b) unlabeled object, (c) completely novel
object in silence
• Recorded EEG gamma band response (index of object
• Gamma band response stronger after the labeled object
relative to unlabeled and novel object
• Conclusion: labels modulate object representation
• But this label training is unlikely to result in long-term word
learning (Horst & Samuelson, 2008), so whether learned language
modulates learned representations remains unclear
• Infants’ looking times reflect what they have learned (Fantz,
1964), so we can use looking times to index differences in
• Test word learning to be sure effect is based on long-term
• This work was supported by the International Centre for Language and
Communicative Development (LuCiD).The support of the Economic and
Social Research Council [ES/L008955/1] is gratefully acknowledged. We
would like to thank the caregivers and infants who made this work
At-home 3D object training
• Caregivers train 10mo infants (N = 24) with 2 novel 3D
objects in 5 minute play session, every day for 7 days.
• Train label for one object only (counterbalanced)
After 1 week: lab-based eyetracking task
1. Familiarization: present infants with 8 images of each item,
interleaved, silent 10s trials
2. Word learning test trial: both objects presented
simultaneously, with label, 12s
Look! A tanzer!
Look at that!
Longer looking to
(beta = 0.082, SE = 0.032,
2(1) = 5.97, p = .014 )
1 2 3 4 5 6 7 8 9 10 11 12
Proportion target looking
Response at 6556 ms
2s postlabel from
(t(22) = 2.00, p = .058)
a b c d e f g h i j k l m n o p q r s t u v w
Change in proportion target looking
Some (but not all)
Introduction Method Results
Two possible mechanisms
• Label activation: labeling during training increases
attention (Baldwin & Markman, 1989), during silent familiarization effect
persists / implicit activation of label increases attention (cf.
Mani & Plunkett. 2010)
• Labels and objects represented separately; labels augment
object representations over experience (Westermann & Mareschal, 2014)
• Novelty: label and object representation become
integrated during training; absence of label during
familiarization causes novelty response
• Labels are features of object representations in the same
way as color/shape/texture (Gliozzi, Mayor, Hu & Plunkett, 2010)
• Computational test of these alternatives: see Capelier-
Mourguy, Twomey & Westermann (2016, August). A
neurocomputational model of the effect of learned labels
on infants’ object representations.
Althaus, N., & Westermann, G. (2016). Labels constructively shape categories in 10-month-
old infants. JECP.
Baldwin, D. A., & Markman, E. M. (1989). Establishing word-object relations: A first step.
Bergelson, E., & Swingley, D. (2012). At 6–9 months, human infants know the meanings of
many common nouns. PNAS.
Fantz, R. L. (1964). Visual experience in infants: Decreased attention familar patterns
relative to novel ones. Science.
Gliga, T., Volein, A., & Csibra, G. (2010). Verbal labels modulate perceptual object
processing in 1-year-old children. J Cognitive Neurosci.
Gliozzi, V., Mayor, J., Hu, J. F., & Plunkett, K. (2009). Labels as features (not names) for
infant categorization: A neurocomputational approach. Cognitive Sci.
Lupyan, G. (2012). Linguistically modulated perception and cognition: the label feedback
hypothesis. Front Psych.
Mani, N., & Plunkett, K. (2010). In the infant’s mind’s ear: Evidence for implicit naming in
18-month-olds. Psych Sci.
Perry, L. K., & Samuelson, L. K. (2011). The shape of the vocabulary predicts the shape of
the bias. Front Psych.
• www.lucid.ac.uk / wp.lancs.ac.uk/westermann-lab/