• change detection task
• fixed-resolution hypothesis
• flexible-resource hypothesis
• hybrid model
• slots + resources model
• visual memory

Visual short-term memory (VSTM) allows for temporary storage of visual information. Until a few years ago the model of visual memory was dichotomous. It was thought to consist of Iconic Memory (IM) , a very high capacity storage of short duration and Visual Working Memory (VWM) with very limited capacity and longer duration.
Recent studies have provided evidence for an additional intermediate form of VSTM that lies between IM and WM: Fragile Visual Short Term Memory (FM) with high capacity and prolonged duration of up to 4 seconds.
Currently there is an ongoing debate about the characterization of the capacity limit of VWM; some researchers argue that VWM has an upper limit of 2-4 representations with a fixed resolution per representation (slot model), while others argue that there is no upper limit in the number of representations, but there is a trade-off between the number of representations and the resolution per representation (resource model).
The present thesis addresses this issue and investigates the capacity and resolution not only concerning VWM, but also Iconic Memory and Fragile VSTM. Our goal is to provide evidence for or against the slot/resource models per memory type. Moreover, we investigate metacognition to determine whether performance was influenced by some kind of blindsight like phenomenon.
We use a change-detection partial-report task with eight oriented bars, where participants have to compare the test array with the memory array indicating whether one bar (indicated by a cue) is rotated or not. To evaluate the resolution at which an item is encoded we use three different angle rotation (30°, 60° or 90°); furthermore, through the use of a precue, we manipulate the amount of attention allocated on the item to-be-remembered. In order to evaluate the metacognition, at the end of each trial, participants have to indicate how much they were sure about their answer.
Our findings support that attention influences both the capacity and the resolution of VWM, suggesting that the resources of this memory type can be distributed asymmetrically; while in FM and IM, attention affects only the capacity, but not the resolution of the encoded representations, suggesting that they may be based on slots with fixed precision. Furthermore our results revealed that for the three memory types, the performance is not influenced by some kind of blindness.