Visual imagery

Mental imagery

Our ability to imagine mental representations that aren’t currently present. Takes place in each sense modality: vision, audition, touch, smell, pain, etc.

Visual imagery

Many philosophers speculated that thought consists of nothing but a sequence of visual images, but things like abstract concepts and verbal rehearsal are thoughts that don’t involve visual images.

2 types of cognitive representations

Modal: representations in the same modality as one of the senses. i.e., Visuospatial sketchpad = spatial representations, Phonological loop = acoustic representations.

Amodal: representations not tied to any sense modality. e.g., propositional networks.

Propositional networks

A proposition is a language-like structure that represents a simple fact, i.e., “the cat is on the mat”. This network is a graphical representation of a set of related propositions: taxonomic hierarchies, part-whole relations, causal relations. For instance, a taxonomic hierarchy is living thing → animal vs. plant → mammal vs. bird → dog vs. cat.

Does cognition sometimes involve visual imagery? or are proposition sufficient to explain all cognitive activity?

There is strong evidence that visual imagery involves spatial representations. Propositional networks do not suffice as an explanation of visual imagery experiments.

Mental scanning (Kosslyn) experiment

Ss asked to view then memorize pictures and when asked about certain picture, to imagine it. They focused on one part and then were queried about another. Focus condition asked Ss to focus on one part of the image, i.e., top, bottom, left, right. Whole condition asked Ss to focus on the whole image. Ss were then asked if the image contains X property.

Results show that RTs increased as the distance in the visual image between focus point and X property increased, this was only for the focus condition though. Imagined visual images exhibit some of the same properties as actual visual images.

Alt explanation to mental scanning

Are some mental scanning effects possible from propositions? Perhaps, time to scan is a function of the number of propositions in a propositional network. I.e., from the motor of the boat → rear deck → cabin → front deck → anchor.

Mental scanning (Kosslyn, Ball, & Reiser) experiment

Ss asked to memorize a map. Ss then asked to mentally scan from one location to the other, e.g., from beach to well.

If we used a visual image, RT from beach to rocks should be longer than from the beach to well. Using a propositional network, RT from beach to rocks should be shorter than from the beach to well.

Results show that RTs were longer as “distance” in image increased, rather than no. of intervening items. This tells us visual image is more potent.

Picture superiority effect

This is the phenomenon where memory for pictures is better than words. Due to 1) distinctiveness and 2) elaborations that produce multiple codes.

Images and distinctiveness

Similar memory items interfere with one another during retrieval. So because images have a large amount of detail, they are very distinct, hence, there is little competition between them, minimizing interference.

Dual-coding hypothesis (Paivio) experiment

Visual stimuli can be elaborated with verbal information, which as we know elaborations allow retrieval via multiple cues. Thus pictures can be stored in two ways: image and verbal/propositional code.

Memory improvement techniques that make use of imagery

1. Keyword technique: associating foreign vocabulary to familiar English words and phrases.

2. Method of loci: attributed to Greek poet. A to-be-memorized list of items are associated with locations on a familiar path. The memorizer constructs an image with includes both new item and familiar location. Shown to yield 70-100% improvement in recall.

Concrete vs. abstract imagery (Paivio & Foth) experiment

Asking if imagery would also help in memorization of abstract material.

Ss asked to study lists of paired nouns that were either concrete or abstract hence the two conditions. They were then asked to link the pair of nouns with a mental image and draw it—imagery condition—or link the pair of nouns with a word or phrase and write it down—verbal condition. Cued recall test of the paired nouns.

Results show that imagery enhances memory of concrete material but it hurts memory of abstract material. For abstract material, semantic elaboration usually better leads to better memory.

Pros and cons of visual memory

It is generally very good as many memory techniques rely on forming mental images. But imagery may not help, and may even hurt memory for abstract material.

Memory for faces (Bahrick et al.) experiment

This told us that while picture memory is very good, memory for faces is extraordinary. Ss picked former classmates from a group of five pictures, and we can see that people’s memory for faces from HS remained >70% up until 47 years after graduation.

Specialization in face recognition (Yin) experiment

There is some evidence that there are mental processes dedicated to the perception of and memory for faces. An example of this specialization is that recognizing rotated faces is more difficult than other objects.

Ss were tasked to recognize 4 types of familiar images including faces, presented either upright or inverted. Results showed that all objects are harder to recognize when inverted but the deficit is especially dramatic for faces. This is due to the fact that faces are recognized holistically, so relations are critical, e.g., the nose is above the mouth.

Prosopagnosia

A condition where people recognize objects but not faces.

Evidence that faces are special

• Prosopagnosia

• Newborns’ preference to view faces

• Certain areas of the brain (the fusiform gyrus) are active during face perception

Spatial knowledge

This is also in semantic memory and tells us that images have parts which have subparts, etc., it appears to have a hierarchical structure. But this is also prone to distortions, i.e., “if you head south from Texas which South American country will you hit?” the answer is none, but people usually choose one anyway.

What do we now know about visual imagery?

It is indeed based on vision-like spatial representations. Visual images are much more likely to be remembered, BUT not when material is abstract, and inverted faces are hard to identify, and spatial knowledge is organized hierarchically; prone to distortions.