Sorea Karimi, Leonie Baumann, Marina Proft, Hannes Rakoczy and Tanya Behne
Affiliation: Georg-August University Göttingen
Category: Psychology
Keywords: modal cognition, modal language, cognitive development, modal reasoning
Date: Wednesday 3rd of September
Time: 17:00
Location: Gen. Henryk Dąbrowski Hall (006)
View the full session: Modal Cognition & Problem Solving
Modal cognition involves the ability to differentiate between what will necessarily occur and what could possibly occur. When an object is dropped down a tube with two exits (Y-tube), the object could come out on the right side but it does not have to come out on the right side. Thus far, research on children’s modal cognition reveals a puzzling land-scape: When faced with such situations as the one described above, which involve mutu-ally exclusive possibilities, children below the age of four have repeatedly been shown to struggle in responding appropriately (e.g., Beck et al., 2006; Leahy, 2023; Mody & Carey, 2016; Redshaw & Suddendorf, 2016; Redshaw et al., 2018; Robinson et al., 2006; Sudden-dorf et al., 2020). In the case of the Y-tube for example, two-year-olds extended their arm only towards one exit when instructed to catch the object, thus preparing for only one of two possibilities, as if it were a necessary outcome (Redshaw & Suddendorf, 2016; Red-shaw et al., 2018). The theoretical account of minimal representation of possibilities attributes chil-dren’s poor performance in the aforementioned studies to a lack of mastery of modal lan-guage (Leahy & Carey, 2020). The authors argue that modal language is a prerequisite for constructing the modal concepts that are needed for proper modal cognition. In support of this claim, Leahy and Žalnieriūnas (2021) found that even four-year-olds often under-stand the modal “have to,” to express possibility rather than necessity. In their study, they presented children with a branching and non-branching slide and asked them whether the marble “can” or “has to” come out of an indicated exit. They did this sequentially for each entrance and exit combination. Their findings are in line with other research, which also identified young children’s tendency to conflate necessity modals with possibility modals, and vice versa (e.g., Dieuleveut et al., 2022; Moore et al., 1990; Öztürk & Papafragou, 2015). The research landscape is puzzling however, because other studies cast doubt on the claim that younger children cannot use modal reasoning specifically due to their non-adult like modal language abilities. Recent studies did find evidence for modal reasoning in children below the age of four (Alderete & Xu, 2023), for example, by reducing demands on action planning and increasing the agentive scope of action (Baumann et al., 2024; Turan-Küçük & Kibbe, 2024). Furthermore, eye-tracking data seems to indicate that 14-month-olds represent multiple possibilities (Cesana-Arlotti et al., 2022) and that four-year-olds differentiate between “must” and “might” the same manner as adults (Moscati et al. 2017). Additionally, Armstrong (2020) observed that even three-year-old children were capable of differentiating between necessity expressions (“will”) and possibility expressions (“might”). Taking all of these findings together, the question about which role modal language might play in modal cognition as well as the question of how young children even understand common modal expressions remain unresolved. Thus, to further investigate young children’s modal language abilities, the present study set out to examine four- to six-year-old’s understanding of necessity modals (“defi-nitely have to”) and possibility modals (“can”). To do this, we devised two new task (selec-tion task and production task) which were embedded in a coin tossing game (adapted from Baumann et al., 2024). Coins were chosen to avoid masking children’s possible com-petences by requiring them to reason about physical trajectories as is the case in the Y-Tube task and slides task. The game context was chosen to keep children engaged and thus support them in demonstrating their assumed modal language competencies. Children were introduced to a dog and a giraffe which were to play a coin-tossing game. While the dog would win on red but lose on yellow and brown, the giraffe would win on yellow but lose on red and brown. In the selection task, one animal arrived and present-ed three differently colored coins to the child and asked them to indicate all the coins where it could or definitely had to win or lose. Then, the other animal brought three new coins and asked the same set of questions (4 trials each, 8 trials in total). In the production task, children crafted the coins themselves based on an instruction manual that the exper-imenter read to them. In four trials, they were instructed to craft coins such that one or both animals can or definitely has to win or lose. Verbal ability was controlled for using the K-ABC II (Kaufman et al., 2005). To analyze the data, a binomial mixed effects model will be fitted with task type and modal expression as the predictors of interest of correct responses in the two tasks. Response patterns in necessity and possibility trials will be compared descriptively to identify whether children treat these modal expressions differently. The project is currently in data collection, which will be finished by June 2025. Preliminary results (n = 37 for the selection task and n = 38 for the production task out of N = 75) suggest that both the modal expression (z = 1.26, p = .007) and the task type (z = -1.40, p = .012) might significantly pre-dict children’s correct responses. Furthermore, children’s response patterns in the selec-tion task but not in the production task seem to align with those reported by Leahy and Žalnieriūnas (2021). These preliminary results could be taken as first indications that the tasks we use to test modal language competence and modal reasoning need closer inves-tigation in order to determine the circumstances under which young children may be bet-ter enabled to exhibit modal language competence.