See more Criminal Appeal 3055/18 Abu Raqaiq v. State of Israel, Verse 5 (August 4, 2020), where it was held as follows:
"Identification is a challenge. On the one hand, experience clearly shows that a witness can identify another person. On the other hand, 'you live from the mouth' of the witness. This means that there are no laboratory conditions through which the testimony can be examined, just as it is possible, for example, to examine DNA evidence. It is plain that a person's thought process and identification mechanisms are complex. Therefore, there are characteristics of such evidence that are not the property of evidence of another kind. For example, it is customary to ask the identifying witness with what degree of confidence he identifies the person he is pointing to as the perpetrator of the offense. This stage of self-esteem is important, but it can also be misleading, because the person quantifies something - the confidence in the testimony - which is nothing more than a kind of personal estimation. There is also another complexity, such as the personal data of the identifying witness: Is there a difference between a young witness and an older person? Other examples are a person who has difficulty seeing as opposed to a witness who does not need glasses, or the degree of lighting and distance between the witness and the suspect. Another factor is the history between the identified and the identifier - whether it is a one-time and initial encounter or a daily acquaintance. Still, as with any testimony, its reliability and weight must be examined according to rules common to all evidence.
In order to approach the work, it seems to me that it would be appropriate to use a double test - internal and external. The initial test is an internal examination of the identification evidence. The second is an external examination. The first will require going into the details of the testimony: the degree of clarity, the existence or absence of contradictions, the content of the testimony, and the extent to which the court believes that the witness describes what he actually saw. After all, until he claimed that he was 30% sure of his testimony, as long as he was 99% sure of his testimony. The court can reject the testimony of an identification witness according to the internal test, such as a determination that the prosecution's witness is lying and was not at all at the place of identification. In such a case, the testimony will not help the prosecution.