Academia.eduAcademia.edu

Figure 1 – uploaded by Kirsty Kitto

See full PDFdownloadDownload figure
We can learn a little about this model through a consideration of the two terms in (5). The first term represents the desired vector b; , weighted by a scalar term that results from taking Retrieval from this list memory is defined by Humphreys et al. [10] to work as follows. First, a test cue xa, is applied to the list memory:

Figure 1 We can learn a little about this model through a consideration of the two terms in (5). The first term represents the desired vector b; , weighted by a scalar term that results from taking Retrieval from this list memory is defined by Humphreys et al. [10] to work as follows. First, a test cue xa, is applied to the list memory:

Related Figures (2)

Fig. 1. A concept w, for example red, is represented in some contextual probe c which takes the form of a choice of basis. This low dimensional representation shows a case of some object being classified by a probe as “red” (|1)) or “not red” (|0)), with the probability of that a “red” judgment being given by |a;|?. In a higher number of dimensions the property of redness will be enclosed by a convex region, and the probabilities lie in a range of values specified by the extent of that region. Here, {|0,) = (0, 1)”, |1,) = (1,0)"} define an orthonormal
Fig. 1. A concept w, for example red, is represented in some contextual probe c which takes the form of a choice of basis. This low dimensional representation shows a case of some object being classified by a probe as “red” (|1)) or “not red” (|0)), with the probability of that a “red” judgment being given by |a;|?. In a higher number of dimensions the property of redness will be enclosed by a convex region, and the probabilities lie in a range of values specified by the extent of that region. Here, {|0,) = (0, 1)”, |1,) = (1,0)"} define an orthonormal
However, in the context represented by the shifted basis in figure 2 the probe returns the desired information with a probability given by P(|1)) = |b,|?. Thus, a search for a Fig. 2. Changing the context of a probe can significantly affect the chances of recall. Once this effect is incorporated into a Matrix-like memory model, a structural information storage system becomes more viable.
However, in the context represented by the shifted basis in figure 2 the probe returns the desired information with a probability given by P(|1)) = |b,|?. Thus, a search for a Fig. 2. Changing the context of a probe can significantly affect the chances of recall. Once this effect is incorporated into a Matrix-like memory model, a structural information storage system becomes more viable.
Related topics:
Computer ArchitectureInformation RetrievalContext ModelingCognitive Models
580 California St., Suite 400
San Francisco, CA, 94104
© 2025 Academia. All rights reserved