Commonality of processes underlying visual and verbal recognition memory.

Memory for visual and verbal material engages widely distributed systems, to a large degree focussed in different hemispheres. It might thus be expected that these disparate neuronal populations should display significantly different characteristics in regard to mnemonic performance. Visual memory, fundamental to all human beings, and whose characteristics are largely shared with macaques, was assayed using unique non-objective, colored images lacking ready verbal description and was contrasted with memory for four-letter non-offensive English words. The effects of memory loading, stimulus duration and long-term test intervals (1-2 weeks) were studied in regard to accuracy and reaction times for recognizing initial versus re-exposure to these two types of items. No effects of memory loading were apparent despite the incrementing memory load in the 240-item, running recognition sessions. Words were better remembered than images, both in the long and short term, but the detailed characteristics of reaction times and accuracy in relation to number of intervening items, and in long-term memory were strikingly similar. Given the wide and well-established disparity in cerebral loci participating in linguistic versus image analysis, these multiple similarities in the pattern of mnemonic performance indicate that the underlying neuronal processes must be comparable for remembering either images or words. Furthermore, the strong link manifested between individual items across a varying number of intervening intervals and added items suggests that a phenomenon highly similar to the "stimulus specific adaptation" (SSA), displayed by units in macaque inferotemporal cortex, occurs for each item to be recognized. Finally, the significant augmentation in accuracy both in short- and long-term memory for images when viewing time permits saccades is explained if each saccade and fixational pause recruits additional neurons into the pool displaying SSA, or its equivalent, for the item being viewed.

Blur identification by residual spectral matching.

The estimation of the point spread function (PSF) for blur identification, often a necessary first step in the restoration of real images, method is presented. The PSF estimate is chosen from a collection of candidate PSFs, which may be constructed using a parametric model or from experimental measurements. The PSF estimate is selected to provide the best match between the restoration residual power spectrum and its expected value, derived under the assumption that the candidate PSF is equal to the true PSF. Several distance measures were studied to determine which one provides the best match. The a priori knowledge required is the noise variance and the original image spectrum. The estimation of these statistics is discussed, and the sensitivity of the method to the estimates is examined analytically and by simulations. The method successfully identified blurs in both synthetically and optically blurred images.

On the accuracy of PSF representation in image restoration.

Point spread function (PSF) models derived from physical optics provide a more accurate representation of real blurs than simpler models based on geometrical optics. However, the physical PSF models do not always result in a significantly better restoration, due to the coarse sampling of the recording device and insufficiently high signal-to-noise ratio (SNR) levels. Low recording resolutions result in aliasing errors in the PSF and suboptimal restorations. A high-resolution representation of the PSF where aliasing errors are minimized is used to obtain improved restorations. The SNR is the parameter which ultimately limits the restoration quality and determines the need for an accurate PSF model. As a rule of thumb, the geometrical PSF can be used in place of the physical PSF without significant loss in restoration quality when the SNR is less than 30 dB.