CompCog: Towards a unified account of when external cues are beneficial or detrimental during memory search

想象一下，你正在谈论一个家庭度假，你去了，但你不能记住每一个细节。有时，有人在谈话中插话，提醒你发生的事情，帮助你记住更多。我们能开发一种自动化的方法来选择提醒并提高某人的记忆力吗？本研究旨在通过考察和模拟“记忆空间”中的提醒物之间的相互作用来研究提醒物是如何影响记忆的。这些模型使我们能够测试不同的理论，并对提醒的效果做出预测。这项研究的发现可能对教育环境中的个体或记忆障碍者设计记忆干预措施有价值。与来自环境的线索是有帮助的直觉相反，大量研究表明，在记忆搜索过程中呈现的线索有时可能是有害的。我们认为，线索的有效性关键取决于他们的内容和时间。文献中提出了各种记忆线索的描述性理论，但它们很难在经验上加以区分，这影响了我们预测不同类型的线索及其时机的有效性的能力。为了解决这些问题，我们将把记忆线索的描述性理论转化为完全透明的计算模型。这种方法使我们能够定量比较这些理论对线索影响的预测，并确定与数据最匹配的模型。然后，该模型可以用于模拟提示内容和时间的所有可能组合的效果，从而能够选择最有效的提示。总体而言，拟议的研究计划有助于统一考虑外部线索何时是有益的或有害的。该奖项反映了NSF的法定使命，并已被认为是值得通过使用基金会的智力价值和更广泛的影响审查标准进行评估的支持。

Imagine you're talking about a family vacation you went on, but you can't remember every detail. Sometimes, someone chimes in during the conversation to remind you of things that happened, helping you remember more. Could we develop an automated way to select reminders and improve someone's memory? This research aims to study how reminders affect memory by examining and modeling their interaction in a 'memory space'. These models allow us to test different theories and make predictions about the effect of reminders. The findings from this research could be valuable in designing memory interventions for individuals in educational settings or those with memory impairments.Contrary to the intuition that cues from the environment are helpful, a substantial body of research demonstrates that the presentation of cues during memory search can sometimes be detrimental. We argue that the effectiveness of cues depends critically on their content and timing. Various descriptive theories of memory cues have been proposed in the literature, but they have been difficult to distinguish empirically, which impacts our ability to predict how effective different kinds of cues and their timing would be. To address these issues, we will translate descriptive theories of memory cues into mathematically-transparent computational models. This approach enables us to quantitatively compare the predictions of these theories regarding the effects of cues, and identify the model that best matches the data. This model can then be used to simulate the effect of all possible combinations of cue content and timing, enabling the selection of the most effective cues. Overall, the proposed research program contributes to a unified account of when external cues are beneficial or detrimental.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.