圖片、文字是體驗型產品網絡評論的兩種主要形式，但關于兩者對消費者影響的研究結論尚存在分歧。本研究基于解釋水平理論，通過系列實驗揭示了圖片、文字評論在不同購買可行性情境下對消費者產品評價的影響，及產品類型(懷舊/非懷舊)的調節作用。結果表明:(1)購買可行性低情境下的消費者受文字評論的影響更大，而購買可行性高情境下的消費者受圖片評論的影響更大(實驗1a、實驗1b和實驗2);(2)購買懷舊產品時，購買可行性低情境下的消費者仍受文字評論的影響更大，而購買可行性高情境下的消費者受兩類評論的影響無差異;購買非懷舊產品時，購買可行性高情境下的消費者仍受圖片評論的影響更大，而購買可行性低情境下的消費者受兩類評論的影響無差異(實驗3)。本研究表明，當圖片、文字評論與購買可行性在解釋水平上一致時，影響力最大，且懷舊/非懷舊產品類型會調節這種效應。本研究還討論了如何在實際中運用此結論。

Both online pictorial and verbal reviews play important roles in consumers’ purchasing decisions. Based on construal level theory(CLT), we studied the effects of pictorial reviews and verbal reviews on consumers’ decision. Words are abstract representations of objects, so verbal reviews are characterized as high construal level; while pictures are concrete representations of objects, pictorial reviews are examples of low construal level. Furthermore, according to CLT, products are represented at a higher construal level by consumers whose purchase feasibility is low, while products are represented at a lower construal level by consumers whose purchase feasibility is high. Thus we hypothesized that consumers whose purchase feasibility is low would be more influenced by verbal reviews, whereas consumers whose purchase feasibility is high would be more influenced by pictorial reviews. In addition, this effect would be moderated by the product types(i.e., nostalgic vs. non nostalgic products). A series of experiments were conducted to test these hypotheses. In Experiment 1a and 1b, reviews of tangible product features were used to study the impact of purchase feasibility and pictorial/verbal reviews on product evaluation. Experiment 1a employed a 2(purchase feasibility: high vs. low) × 2(reviews: pictorial vs. verbal) between-subjects design. Experiment 1b employed the same design as experiment 1a, but used scenarios which were more close to real life purchasing experience. Results of Experiment 1a and 1b showed that when purchase feasibility was low, participants were more likely to be persuaded by verbal reviews, as compared with pictorial reviews. By contrast, when purchase feasibility was high, participants were more likely to be influenced by pictorial reviews than verbal reviews. These results provided evidence for congruency effect between purchase feasibility and reviews forms. In Experiment 2, we studied reviews of intangible product features of hotels, and changed purchase feasibility conditions. Results of Experiment 2 replicated the congruency effect found in Experiment 1 with the reviews of intangible product features. In Experiment 3, we further studied the moderating role of product types(i.e., nostalgic vs. non nostalgic products) in the relationship between purchase feasibility and reviews forms. We adopted a 2(reviews: pictorial vs. verbal) × 2(purchase feasibility: high vs. low) × 2(product type: nostalgic product vs. non-nostalgic product) between-subjects in Experiment 3. Results showed that the congruency effect between purchase feasibility and pictorial/verbal reviews was moderated by product types: when purchasing nostalgia products, consumers whose purchase feasibility was low were more influenced by verbal reviews than by pictorial reviews, but the difference between the two types of reviews was not significant when purchase feasibility was high; when purchasing non-nostalgia products, consumers whose purchase feasibility was high were more influenced by pictorial reviews than by verbal reviews, but the difference between the two types of reviews was not significant when purchase feasibility was low. The current research contributes to the understanding of how pictorial reviews differ from verbal reviews in affecting consumers’ product evaluation. Moreover, our findings also imply that displaying verbal/pictorial online reviews according to consumers’ purchase feasibility would enhance their purchasing intention, and that the effectiveness of marketing strategies with regard to nostalgia/non-nostalgic products is determined by different factors such as purchasing feasibility.