Orrantia Josetxu, Rodriguez Laura, Vicente Santiago
Universidad de Salamanca, Salamanca, Spain.
Q J Exp Psychol (Hove). 2010 Feb;63(2):310-9. doi: 10.1080/17470210902903020. Epub 2009 May 13.
Studies of mental arithmetic have shown that adults solve simple arithmetic problems by retrieving an answer automatically from a network of stored associations. However, most studies have been limited to single-digit addition and multiplication problems. In this article, we examine whether retrieval is also automatic in the context of more complex arithmetic tasks, such as arithmetic word problems. To test this hypothesis, we used a priming procedure with a target-naming task, in which the primes were the numbers included in two sentences containing the numerical information of an arithmetic word problem (e.g., 3 and 2 in "Joe had 3 marbles. Then Tom gave him 2 marbles"), and the targets were either congruent (e.g., 5) or incongruent (e.g., 8) with the prime. A neutral prime was also used replacing the numbers of the problem by capital letters (e.g., X and Y). Manipulating the relationship between the prime and the target and the duration of time that separates these two events, the overall results revealed shorter times in naming the congruent target than in a neutral condition and longer times in naming the incongruent target, even though mental arithmetic was completely irrelevant to the task. These results support the notion that automaticity of arithmetic-fact retrieval is not limited to simple addition, but it is also possible in other tasks, such as arithmetic word problems, which demand more cognitive resources than single-digit addition.
心算研究表明，成年人通过从存储关联网络中自动检索答案来解决简单的算术问题。然而，大多数研究仅限于个位数加法和乘法问题。在本文中，我们研究了在更复杂的算术任务（如算术应用题）中，检索是否也是自动的。为了验证这一假设，我们在目标命名任务中使用了启动程序，其中启动刺激是包含算术应用题数字信息的两个句子中出现的数字（例如，“乔有3颗弹珠。然后汤姆又给了他2颗弹珠”中的3和2），目标刺激与启动刺激要么一致（例如，5），要么不一致（例如，8）。还使用了中性启动刺激，即用大写字母替换问题中的数字（例如，X和Y）。通过操纵启动刺激与目标刺激之间的关系以及这两个事件之间的时间间隔，总体结果显示，即使心算与任务完全无关，命名一致目标的时间也比中性条件下短，命名不一致目标的时间更长。这些结果支持了这样一种观点，即算术事实检索的自动性不仅限于简单加法，在其他任务（如算术应用题）中也有可能，而算术应用题比个位数加法需要更多的认知资源。
