Are parity and magnitude status of Arabic digits processed automatically? An EEG study using the fast periodic visual stimulation

Presentation (2017, February 22)

Many studies have shown that humans can easily extract numerical characteristics of single digits such as numerical magnitude and parity status. We investigated whether spontaneous processing of magnitude ... [more ▼]

Many studies have shown that humans can easily extract numerical characteristics of single digits such as numerical magnitude and parity status. We investigated whether spontaneous processing of magnitude or parity status can be observed when participants are passively presented with sequences of briefly displayed Arabic digits. We assessed the parity processing by presenting seven odd digit numbers followed by one even digit (and reverse) with a sinusoidal contrast modulation at a frequency of 10HZ in one-minute sequences. The same paradigm and frequencies were used to investigate magnitude processing (i.e. seven digits smaller than five followed by one digit larger than five; and reverse) and control condition (i.e. sequence of 1-4-6-9 followed by 2-3-7 or 8). We observed a strong EEG activation on right parietal electrodes and a weaker activation on left parietal electrodes in all conditions. Left and right activations were stronger in the parity than in the other conditions, reflecting an automatic retrieval of parity information conveyed by the Arabic digits. The weaker activation during the magnitude task could reflect a more complicated access of the information corresponding to magnitude status. Right activations during the control task could be due to the fact that subjects can quickly learn to categorize numbers arbitrarily. These neuronal activation patterns are consistent with the neuro-imaging literature describing the localization of basic numerical processing. Our findings indicate that magnitude and parity status are extracted automatically from Arabic digits, even when numerical stimuli are presented without instructions at a high presentation rate. [less ▲]

THE IMPACT OF LANGUAGE BACKGROUND ON BASIC MATH COMPETENCE

Presentation (2016, February 18)

German number word system inverts units and tens compared to the Arabic notation. This is not the case in French, which is more transparent regarding the Arabic number code. Evidence indicates that the ... [more ▼]

German number word system inverts units and tens compared to the Arabic notation. This is not the case in French, which is more transparent regarding the Arabic number code. Evidence indicates that the linguistic structure of number words can facilitate or impede numerical development (Zuber, Pixner, & Moeller, 2009). Moreover, in transcoding tasks more mistakes are made in non-transparent compared to transparent languages (Imbo, Vanden Bulcke, De Brauwer, & Fias, 2014). We used a new paradigm of transcoding task in which 28 French-speaking (FR) and 19 German-speaking (GE) 4th grade children had to listen two digits numbers. The new thing was that we manipulate the order of appearance of the units and the tens of the number in three conditions: Units-First (UF), Tens-First (TF) and Simultaneous (S). Then, the subjects had to choose the heard number among four numbers presented on the computer screen. Results sows that GE are globally slower than FR (F(1,45) = 3.95, p = .053). The largest difference was observed for the TF: (t(45) = -3.729, p = .001). Moreover, when the order of the number appearance was congruent with the number word system, the transcoding was faster in both languages. For GE the S condition was slower than TF condition (F(2,36) = 6.918, p = .008) and than UF condition (F(2,36) = 6.918, p = .003.). For FR, the TF was faster than S (F(2,54) = 69.419, p < .001) and UF (F(2,54) = 69.419, p < .001). All these data indicate that language structure qualitatively impacts on basic numerical tasks. [less ▲]