Many naming and narrative tests  use images as a support to elicit the production of words and phrases. Other tests use physical objects. Why? The most accredited theories on language processing agree on the existence of a single semantic center (it would, in fact, be uneconomical to think that there is a semantic center for the images we see and another for the words we hear), but at the same time they do not believe that the different input channels access them with the same ease.
For some it may seem trivial, for example, that the image of a hammer can guarantee faster access to the characteristics of the hammer than the word "hammer" (the latter being, like all words in our language, arbitrary); however, we might be led to think that both the image of the hammer and the word "hammer" are just gods access points to the idea of the hammer, and therefore regardless of the channel, the semantic characteristics are activated only by the idea of the hammer. Some studies, including the 1975 Potter historical one  have shown that this is not the case, and have done so by showing different naming times depending on the different channel used.
If, in fact, from the second year of primary school onwards, the reading of a word is faster than the naming of its image, it is also true that the attribution of an element (for example, a table) to a category, is more quick when the object is presented as an image and not as a written word. Many authors speak in this sense of privileged access (direct link between stimulus and meaning) e privileged relationship (connection between the structural aspects of the stimulus and the semantic properties connected to its action) of objects - and images - with respect to semantic characteristics.
What are the privileged accesses on which we have the most evidence?
- Objects have privileged access to semantic memory with respect to words 
- Words have a privileged access to phonological characteristics compared to images 
- In particular, among all the semantic aspects, objects have privileged access to the action to be performed 
In more recent years, with the emergence of "embodied" theories (see, among others, Damasio) more refined experiments have been carried out on semantic activation related to the objects we use. In a very recent study  people were asked to respond (by moving a lever forward or backward) after observing images, deciding whether:
- Experiment A: the object was used towards the body (ex: toothbrush) or away from it (ex: hammer)
- Experiment B: The object was handmade or was it natural
The authors went to observe the congruence effect, or if the participants were quicker to respond when there was a congruence between the type of object and the movement of the lever (eg: toothbrush, or object to use on me - lever downwards). If, in the first case, the presence of the congruence effect was almost taken for granted, it was interesting to note that, even in experiment B, where the question was not related to the use towards oneself or away from oneself, the congruence effect is it has occurred anyway. In a certain sense, the image of the object "activates" the action in a latent way even if the question we are asked is not related to its use.
Privileged access, therefore, seems to be a phenomenon that does not only concern the visual characteristics of the object, but also our corporeality and the way we interact with it.
 Scotto di Tella G, Ruotolo F, Ruggiero G, Iachini T, Bartolo A. Towards and away from the body: The relevance of the direction of use in the coding of object-related actions. Quarterly Journal of Experimental Psychology. 2021;74(7):1225-1233.