Sixteen healthy subjects serially added a series of single-digit numbers during the MO-numeral task and silently rehearsed a memorized seven-digit number during the VR task.
The rostral and dorsal part of the lateral premotor cortex (PMdr) showed significantly greater activity during the MO-numeral task than the VR task whereas its caudal part (PMdc) was similarly active during the two tasks.
Therefore, given a choice between R and 21 dots, the monkeys should slightly prefer R. We tested this prediction using the task illustrated in Fig. 3e in which the monkeys chose between a dot array and a numeral: the same four monkeys were tested for 4 weeks on the dots-vs-numerals task.
The first thematic set consisted of two tests measuring children's number knowledge: "Say as many numbers as you can" (number recitation) and a number recognition task (Arabic numerals from 1 to 10 were shown in random order).
The number comparison task assesses the capacity to order numerosities by magnitude and understand the numerals.
No activation above control condition, arithmetic tasks performed with Roman numerals, was found during mental arithmetic tasks performed with Arabic numerals in either AG region [ 21].
Both MO-numeral and VR tasks revealed activity within the lateral premotor cortex among others.
Task-irrelevant Arabic numerals (1, 4, 6, 9) appeared simultaneously at the same location as a colored target.
Tasks tapping different cognitive functions assigned to asemantic (AS) vs. semantic (S) processing (P) and separated into different types of material being compared (numerals vs. letters, numerals vs. words); correspondingly tasks are grouped into three task groups (TG): I, II, and III.
The results of the study posed a serious challenge to the hypothesized single amodal numerosity representation, in that different regions were activated during tasks involving dots and numerals.
Her task was to select each numeral in the correct order.
