Abstract:  An important aspect of human cognition is that performance depends on the strategies that people use in a wide range of cognitive domains. In the field of arithmetic cognition, the performance of strategy utilization is influenced by many factors among which central executive functions (central executive functions, EFs) are involved and play critical roles. In previous studies, researchers tended to focus on the impact of central executive load on the use of arithmetic strategy in children or adults with typically development (TD) of mathematical ability, whereas children with mathematics difficulties (MD) had been out of consideration. Consequently, the present study used the choice/no-choice paradigm to investigate the influence of central executive load on the strategy utilization of children with MD during arithmetic processing. Seventy-four sixth graders consisting of 36 MD and 38 TD were selected in accordance with previous studies strictly. All participants were asked to finish a two-digit addition computational estimation task with rounding-up or rounding-down strategy and a secondary task at the same time. The task consisted of two parts, strategy execution and strategy choice. Meanwhile, we manipulated varying central executive load as a secondary task by requiring participants to memorize five (high central executive load) or three (low central executive load) digits presented randomly in sequence at the beginning of each trial and ranked them with descending order or no digit (no central executive load). Results showed that: (1) In terms of the strategy executive, the accuracy rate of MD was significantly lower than TD under the three central executive load conditions. The reaction time of MD was significantly longer than TD under both “no” and low central executive load conditions but not for high central executive loaded condition. With the increasing of central executive load, the MD had distinct manifestation compared with TD on strategy executive; (2) In the case of strategy selection, the accuracy rate of MD was significantly lower than TD and the reaction time of MD was significantly longer than that of TD under three central executive loaded conditions. Compared with TD, the strategy selection of MD was more heavily influenced by the levels of central executive load. Likewise, the performance of MD was influenced by growing central executive load on strategy selection other than TD; (3) For the adaptivity of strategy choice, the higher of the central executive load level was, the worse the adaptivity of strategy choice of participants became, and the adaptivity of strategy choice of MD was significantly worse than TD under all of three central executive load conditions. These findings have important theoretical and practical significance for the profound understanding of the potential mechanism of MD with worse strategy utilization. In conclusion, the strategy utilization of MD is worse than TD during arithmetic processing. Though strategy utilization of both the MD and TD is getting worse as a function of the levels of central executive load, MD are more heavily influenced by central executive load. The adaptivity of strategy choice of two groups is affected by the strength of central executive load and the adaptivity of strategy choice of MD is always worse than TD. Moreover, the present study contributes to explore how the central executive load influences the process of strategy utilization clearly. Combining with event-related potential (ERP) as well as functional magnetic resonance imaging technology (fMRI), future research should further explore the changes of individuals’ ERP components or brain regions activated under different levels of central executive load.

Key words: strategy utilization, central executive load, inhibitory control, mathematics difficulties, computational estimation