Memory distortion and the need for systematic recall - Use a database

Good decision-making is often based on our ability to gather past data quickly in order to assess the likelihood of a current situation behaving like the past. If an investor has good "memory", then his ability to recall the past can be effectively employed, but if there are memory distortions or lapses then the recall of the past can harm decisions.

Early simple modeling of memories viewed the mind like a file cabinet where a person will just grab data from internal files as needed. There may some data in long-term memory in the back of the file cabinet while other information is easily accessible. These memories have been encoded and it is just a matter of retrieving them from storage. This analogy may suggest cluttering and data lost or misclassified, but it generally worked. 

However, memories are much more complex than presented with the file metaphor. The memories of humans are filled with distortions and flaws that create biased that need to be adjusted. Memories are tied to narratives and mixed with different levels of vividness and encoding. Memory distortions could be called behavior biases, but the issue is more fundamental.  

One of the key roles in systematic management is data storage and retrieval. If you have the data and can manipulate it into a form that is useful, better decisions will be made. If you don't have the data, inferences will be flawed even if the decision process is effectively designed. Storage and use of data can be better done by computers that manage data without memory or retrieval glitches.

An investor could ask a question of how inflation compares with the past. The investor could then use memory to describe market sentiment during the last period when inflation was rising. A memory could be pulled forward or recalled, but the entire process can be more effectively developed or described through data management techniques. Memories will be subject to distortion, vividness biases, or just simple forgetfulness. This is less possible with good data management. The problem will not hinge on quality or quantity of data but on the manipulation or interpretation of events. Disciplined management with good data should be superior to discretionary management with memory.