Loading...
|
Please use this identifier to cite or link to this item:
https://nccur.lib.nccu.edu.tw/handle/140.119/157126
|
| Title: | 從類別結構、認知處理及性格探討分類策略的個別差異
Investigation of Individual Differences in Categorization Strategy: Category Structure, Cognitive Processing, and Personality |
| Authors: | 楊立行 |
| Contributors: | 心理系 |
| Keywords: | 分類策略;個別差異;類別學習
Individual Difference;Categorization Strategy;Category Learning |
| Date: | 2020-03 |
| Issue Date: | 2025-05-28 14:06:52 (UTC+8) |
| Abstract: | 在類別學習作業中，常於學習表現和分類策略上，觀察到個別差異。然而，一直到一系列探討類別是由單系統或多系統進行學習之前，都沒有研究特別以類別學習作業中的個別差異為主要研究對象。根據那一系列研究，目前證據明顯地支持，工作記憶廣度愈大，學習的表現也愈好；但，仍然沒有研究說明，分類策略上的個別差異是如何形成的？本研究針對此一議題，由類別結構的歧義性、認知處理上的個別差異，以及個人性格面進行探討。在類別結構方面，實驗一假設並檢驗，當人們發現除了相似性之外，也可以根據類別內刺激組型（configuration of features）預測類別時，便會形成不同的分類策略。實驗二更進一步假設並檢驗，刺激組型還可以用於預測應使用何種分類策略進行分類。實驗三則假設並檢驗，經由刺激組型啟動的分類策略相較相似性判斷的分類策略，於判斷時更為緩慢也較不明確。實驗四除了類別學習之外，更測量參與者的場地依賴傾向、工作記憶廣度、問題解決能力以及認知需求。並利用結構方程模型建立這些心理變項與分類策略之間的相關性。此外，也將使用階層式貝氏模擬架構，以當代類別學習模型適配分類資料，進行參數估計。同樣，利用結構方程模型，建立合適描述上述心理變項與這些參數之間關係的模型。
One characteristic of human category learning is the clear individual differences in terms of the learning performance and the categorization strategy revealed in the transfer phase. However, the issue of how individual differences occur in category learning has not been seriously addressed until a series of studies for clarifying the debate on whether category learning is achieved via single system or multiple systems as assumed in the COVIS model. Regardless of the details, the current evident seems to support that the larger the working memory capacity, the better the performance of learning categories will be. As for how the individual differences in the categorization strategy occur, there has not been research addressing this issue. Therefore, the aim of this study is to investigate the factors for such differences to occur with respect to the characteristic of category structure, the individual difference in cognitive processing, and personality. For the category structure, it is hypothesized that the category structure should be designed to have multiple definitions, namely ambiguous category structure. Also, the key point for the ambiguous structure to induce different categorization strategies is hypothesized to be relevant to the attention toward the category membership represented as the configuration of features. Therefore, Experiment 1 uses the category structure of Conaway and Kurtz (2015) as basis to examine this hypothesis. Also, it is hypothesized that the category membership can be used as a reference for not only predicting category labels, but also deciding which categorization strategy should be used. This hypothesis is examined in Experiment 2. The different strategies assumed in this study always include the similarity strategy (or Proximity strategy) and the strategy driven by category membership. The similarity strategy is thought to be automatic and implicit, whereas the category-membership-driven strategy is thought to be slow and perhaps a temporary heuristic for solving the current categorization problem. Thus, Experiment 3 is designed to examine whether the participant adopting the category-membership-driven strategy would have a larger certainty about their response. In Experiment 4, in addition to examining participants’ categorization strategy, the other human factors will be examined also. For the individual differences in cognitive processing, it is planned to test participants’ tendency of perceiving objects in a filed-dependent way using the framed-line test, participants’ WMC using the working memory battery for MATLAB, participants’ problem-solving ability using the category test in the famous Halstead-Reitan Neuropsychological Test Battery, and the participants’ attitude toward thinking using the need for cognition scale. All participants in Experiment 4 will be asked to do two category learning tasks as well. The categorization performance will be correlated with the above measures using SEM in order to understand the human factor for the individual differences in categorization strategy. For all experiment data, hierarchical Bayesian modeling will be conducted to evaluate the capacity of current models to account for the individual differences in categorization strategy. This is a two-year study. The first two experiments will be completed in the first year. The last two experiments will be finished in the second year. |
| Relation: | 科技部, MOST106-2410-H004-069-MY2, 106.08-108.07 |
| Data Type: | report |
| Appears in Collections: | [心理學系] 國科會研究計畫
|
Files in This Item:
| File |
Description |
Size | Format | |
|---|
| index.html | | 0Kb | HTML | 20 | View/Open |
|
All items in 政大典藏 are protected by copyright, with all rights reserved.
|
