以多組件工作記憶模型探討卡通代言人物圖像和語文資訊的編碼與回憶

Abstract

在傳播領域中，電視、電影、電腦、甚至行動裝置等的多媒體內容常常包含跨感官的訊息。其中，電視廣告除了會搭配演奏歌曲外，近年來越來越多電視廣告搭配包含廣告標語的廣告短歌。觀察這類搭配廣告短歌的電視廣告，發現多數都以企業或商品的卡通吉祥物作為廣告代言人或廣告商品。

本研究試圖以多組件工作記憶模式來探討這類視聽廣告的認知處理方式。多組件工作記憶模式認為，人們透過語音迴路和視覺空間模板來編碼語文和圖像資訊，而情節緩衝器可以暫存圖文整合資訊。根據該模式，當人們進行預載的雙重任務時，由於中央執行器提供的注意力資源有限，故雙重任務會比單獨任務表現差。此外，視覺或聽覺預載不會影響人們回憶圖文整合資訊，然而視聽預載會干擾情節緩衝器，因此會影響圖文整和資訊的回憶表現。

本研究實驗一、二以卡通人物作為刺激物，討論不同預載情境下（無預載/聽覺預載/視覺預載/視聽預載）主要任務的記憶表現。由於實驗一未能有效操弄，實驗二改變預載方式再次進行。實驗二結果雖未達統計顯著，仍然可觀察到幾點發現。首先，三組不同呈現方式的卡通人物再認正確率都接近100%，顯示實驗參與者即使受到視覺預載、聽覺預載或視聽預載干擾，仍能順利編碼與提取卡通人物之記憶。再認反應時間的趨勢則提供兩種可能的解釋：（1）情節緩衝器可能不存在，圖像和語文分別由不同組件處理，並有對應的參照連結；（2）情節緩衝器存在，然而人們根據不同認知風格轉換或維持單維度的編碼，使得訊息最終仍由視覺空間模板或語音迴路單獨處理。至於描述性雖沒有造成顯著的記憶成效，但本研究發現聽覺呈現的卡通人物，其預載類型與描述有交互作用，顯示描述性會影響工作記憶使用不同組建來編碼。最後，實驗三則以真實廣告作為刺激物，結果顯示卡通人物代言類型的廣告搭配廣告短歌具有豐富的提取線索，故廣告內容的提取表現較佳。

The TV commercials sometimes use jingle which use advertising slogans as lyrics in recent years. Moreover, we found the TV commercials with jingle usually make cartoon character as advertising spokesman. The study exploring encoding and recall of verbal and imagery integrated information with cartoon spokes-character advertisement from a perspective of multi-component working memory model.

According to the model, people encode and maintain verbal information through phonological loop and image information through visuo-spatial sketchpad. In addition, the episodic buffer can temporarily maintain integrated information. The study use preload task to figure out how working memory operation. Because the resource of central executive limited, the performance of dual task would be worse than the performance of single task. we predicate audial and visual preload won’t interfere maintain of integrated information. On the other hand, audio-visual preload will interfere maintain of integrated information.

Experiment I and II use cartoon characters as stimulus, exploring their recall performance between different conditions. Because experiment I failed to effectively preload, experiment II change the preload of manipulation. Our study propose two possible explanation for our result: (1) Episodic buffer do not exist. Verbal and image cartoon character information have referential connection. People use separate working memory component to encode verbal and image information; (2) Episodic buffer do exist. People have different cognitive styles, so they will maintain or transfer code of massage. In the other words, they just use phonological loop or visuo-spatial sketchpad to encode. Moreover, we found interaction between preload and description. The result show that The description will affect working memory using a different component to encode.

Finally, experiment III use real TV commercials as stimulus. Our result found that the performance of the TV commercials use cartoon characters as spokesman and jingle is much better. We believe that that kind of TV commercials have more retrieval cues to help recall performance.