There is an even subtler definition regarding ‘shared properties’ of iconic signs. It comes from Pierce who viewed a sign as iconic when it “may represent its object mainly by its similarity”. (cited Eco, p 195) A definition that depends on the notion of similitude, which has ‘scientific status’.
Similitude is less imprecise than shared properties and it tends to include precise parameters that are relevant to that being defined and disregards others as irrelevant. This selected use is clearly seen in geometry, which ‘roughly speaking, defines similitude as likeness between two figures in all but size.
[OS Eco is saying similitude is subtler, but less imprecise than shared properties? It seems contradictory.]
Acceptance of these selective parameters, Eco claims, requires a “certain training”. He uses an example of someone being asked if a miniaturized model of a pyramid and the Cheops Pyramid are the same. Common sense would conclude ‘no’, but “…only after a certain amount of training would my naive interlocutor be able to realize that I am looking for a ‘geometric’ similarity.” (p196)
Returning to the example of the line drawing of the horse, recognition of the line as a horse is a cultural decision that requires training. Therefore, Eco concludes, “Similitude is produced and must be learned“. (Gibson 1966 cited p 200).
[OS It seems the answer to the supposed contradiction is that subtlety does not suggest precision or its opposite: imprecision.]