DS0804 -

Thinking without Words: Abstract Combinatorial Thinking in Infants and Children – TACTIC

Submission summary

The human adult mind is characterized by the cognitive capacity to combine information in a productive way, which sustains language, reasoning, learning and innovation. Little is known about the developmental origins of these abilities, and how they relate to the acquisition of the lexicon. On one extreme view, the human mind is innately equipped with the cognitive capacities to combine information and produce all the representations that form our thoughts (Fodor, 1975). These capacities would predate and support language acquisition. Alternatively, without words, thinking could be extremely impoverished (Bermudez, 2007). On this view, words may constitute a unique medium for two uniquely human abilities: (1) abstract thinking, i.e., representing concepts that cannot be directly perceived through our senses such as freedom or atom, and (2) combinatorial thinking, i.e., merging representations to create a novel concept such as stone lion, expensive cookie, expensive house or not red.
TACTIC aims at characterizing the pre-lexical mind by investigating abstract combinatorial thinking in infancy and childhood and asking what role the acquisition of words plays in the development of that ability. TACTIC will focus on two case studies: negation, and the abstract relational concepts same and different.
First, negation, as a combinatorial logical operator, combines with its argument (e.g., red) to produce a novel concept (not red). We will begin by identifying potential precursors of negation in young infants. For instance, if we ask infants to choose repeatedly between two objects, a ball and something else, can infants learn to always take the object that is not a ball, i.e. learning to avoid the ball? We will investigate whether such negative (avoidance) attitude can be used to build a representation that excludes the ball, such as, “all the possible objects, but the ball”. Using eye-tracking anticipatory looking paradigms, we will also investigate infants’ learning of exceptions to a rule, and the role that inhibition plays in these processes.
Second, the concepts same and different are abstract relations that can apply to any modality and domain of cognition: sounds and images, objects and agents, etc. Intuitively, different can be represented as not same. Therefore, the abstract representation of different in infancy may imply the ability to compute negation. In contrast, its absence may reveal the representational limits of infant cognition. There is a large literature in comparative psychology (see Thompson & Oden, 1996 for review) and a smaller body of work on human infants that suggest that both pre-lexical infants and non-linguistic animals represent the relation same, while it is not clear whether and how they represent the relation different (Hochmann, Mody & Carey, 2016). We will develop novel paradigms to study infants’ ability to represent the relation different, using pupillometry and electroencephalography. Furthermore, we will propose that infants’ representation of the relation same relies on representations of variables and a duplication operation, which turns a variable X into {X X}. In this framework, the representation of different would be {X Y}, plus an additional constraint that X and Y cannot take the same value. This format of representation thus accounts for the greater difficulty to represent different. We will experimentally address the predictions drawn from that hypothesis.
Finally, we will explore the role that the acquisition of single words (“pas”, “même”, “différent” in French) may play in the development of abstract combinatorial thinking. In our working hypothesis, combinatorial representations require limited precursor representations (such as {X X} and the avoidance rule) and words, which are necessary to overcome a number of cognitive limitations.

Project coordination

Jean-Rémy Hochmann (Laboratoire sur le Langage, le Cerveau et la Cognition)

The author of this summary is the project coordinator, who is responsible for the content of this summary. The ANR declines any responsibility as for its contents.


L2C2-CNRS RHONE AUVERGNE Laboratoire sur le Langage, le Cerveau et la Cognition

Help of the ANR 209,199 euros
Beginning and duration of the scientific project: January 2017 - 48 Months

Useful links

Explorez notre base de projets financés



ANR makes available its datasets on funded projects, click here to find more.

Sign up for the latest news:
Subscribe to our newsletter