随着互联网的普及和电子商务的发展，人们能更方便地获取自己需要的各种信息，越来越多的产品也能够通过网购获得。在信息和产品高度丰富的情况下，消费者的选择范围大大扩大了。但为了在这样宽广的范围内选择满意的产品，消费者却常常要付出比以前更大的认知努力。研究者发现，当待选项目过多而难以评估优劣时，个体可能会放弃决策。这对产品提供者还是消费者而言，无疑都是不利的。在这样的情况下，决策支持系统（Decision Support System）应运而生。个性化推荐系统即为决策支持系统的最新实例，目前也受到了越来越多的关注。 前人研究发现，产品类别和认知需求都会影响用户在选择产品时的认知加工过程，进而影响决策的结果。而系统与用户交流的重要媒介——对推荐结果的解释，却少有研究涉及。本研究探讨了影响用户决策过程和结果的三个变量，即对推荐结果的解释、被推荐产品的性质、用户的认知需求对推荐效果的影响。结果显示，解释方式对决策难度、购买意向和系统使用意向均没有显著影响，而产品类别和认知需求在对系统的使用意向的影响上存在交互作用。讨论认为，解释方式的不显著作用的产生可能是因为研究中推荐列表在用户决策中产生的作用较小。而产品类别和认知需求的对系统使用意向则符合理论推导的结果。论文最后依据研究结果，对个性化推荐的实践提出了建议。
With the rapid development of e-commerce, consumers can access large amount of information of more and more products in online shopping. Consumers’ choices are greatly expanded, but they must pay much more effort than the past to search products, process information, and make decisions. Researchers have warned the possibility that when facing too many options, consumers may postpone purchase decision- they simply choose “not to choose”. The result is unfavorable for both product providers and consumers. One type of online decision support system, the personalized recommendation system was introduced to solve the problem.. Previous researches have found that product type and need for cognition will affect decision-making processes, together with the result of decison-making. But researchers have long ignored an important factor in the system-user interaction process - the explanation of recommendation. Though personalised recommendation systems in use provided various types of explanations, few studies have explored how the explanation could affect recommendation outcomes. This study investigated the impact of three factors - the explanations provided when recommending, type of the recommended product, and consumers’ need for cogniton - on purchase decision making process and outcome. Results showed that explanations did not affect dependent variables, and intention to use the system could be affected by product type and need for cognition. The insignificant impact of explanations may be due to the ralative small amount of information in the explanation. . Based on the findings, suggestions for application and future studies were provided in the last chapter of the paper.