The Role of Surface and Structural Similarity in Analogical Reasoning Aging: Based on the Problem-Solving Paradigm
LI Meijia,ZHUANG Danqi,PENG Huamao
DOI: https://doi.org/10.3724/sp.j.1041.2018.01282
2018-01-01
Acta Psychologica Sinica
Abstract:Analogical reasoning is a major form of abstract logical thinking. It refers to the transferring process of knowledge from one situation or field to another. To transfer successfully when facing new problems, it is important for the elderly to retrieve the appropriate sources of knowledge as quickly as possible. Additionally, the elderly rich in life experience are supposed to be the embodiment of wisdom, which is also helpful. Previous research about the elderly’s analogical reasoning mainly focused on the encoding stage, and, in most cases, the task paradigm was rigorous laboratory study. The task paradigm may have generated outcomes against participants’ reactions in natural settings, causing reduced ecological validity and inability to present their real competence. The present study was based on the problem-solving paradigm rather than a laboratory-based experiment. We aimed to figure out whether the elderly retrograde in the retrieval of source problems and whether there exist age differences in the usage of surface information and structural information. The encoding quality between the old and young participants was ensured to be identical. Twenty-nine young adults (age: M = 21.86, SD = 1.43) and thirty-two old adults (age: M = 69.09, SD = 5.11) participated in the experiment. This study used a mixed design in which age group (old/young) was a between-subjects variable and surface similarity (high/low) and structural similarity (high/low) were within-subjects variables. There were two phases included in the whole procedure. During the learning phase, participants learnt twenty source stories and summarized the higher-order relation of each. Three days later, twenty other target stories were presented in which participants were to figure out how to solve the problems with the hints they recalled from the source stories learnt three days ago. The performance of reasoning (quality of solving the target problem according to the source problem) and retrieving (quantity of surface and structural information retrieved) were the dependent variables. The results showed that the elderly experienced analogical reasoning aging. To be specific, they relied more on the surface feature similarity than did the younger adults. As for surface and structural information, the more the information was provided, the better was their reasoning. It was beneficial to both age groups in reasoning when the surface and structural similarity increased, which indicated that the reasoning of the elderly and young people could benefit from better retrieval to the source problems. As for interaction effect, for the elderly in lack of surface information, increasing structural similarity would not enhance their reasoning quality. When there was enough surface and structural information, they performed the best. It implied that the elderly could benefit from structural similarity only on the condition that enough surface information was given. When retrieving information from source problems, surface feature exerted an across-information-type promotion effect on retrieval quality. However, the increase in structural information could only help the elderly to improve their retrieval in structural information. As for the younger adults, structural features and surface features were conducive to their retrieval of homogeneous information. With regard to the relation between retrieval quality and reasoning performance, the correlation was significant in both age groups. It enlightened us that people who cannot reason correctly were probably poor in retrieval. To improve the performance of the elderly in analogical reasoning, we can increase the transmission of surface features, such as designing mobile phone apps with similar interfaces, which can reduce their learning cost. Further studies should focus on designing more realistic contexts as well as delving into the process of how far transfer of learning happens.