No abstract

Mental imagery is a quasi-perceptual experience which resembles perceptual experience, but occurring without (appropriate) external stimuli. It is a form of mental representation and is often considered centrally involved in visuo-spatial reasoning and inventive and creative thought. Although imagery ability is assumed to be functionally independent of verbal systems, it is still considered to interact with verbal representations, enabling objects to be named and names to evoke images. In literature, most measurement tools for evaluating imagery capacity are self-report instruments focusing on differences in individuals. In the present work, we applied a Mental Imagery Scale (MIS) to mental images derived from verbal descriptions in order to assess the structural features of such mental representations. This is a key theme for those disciplines which need to turn objects and representations into words and vice versa, such as art or architectural didactics. To this aim, an MIS questionnaire was administered to 262 participants. The questionnaire, originally consisting of a 33-item 5-step Likert scale, was reduced to 28 items covering six areas: (1) Image Formation Speed, (2) Permanence/Stability, (3) Dimensions, (4) Level of Detail/Grain, (5) Distance and (6) Depth of Field or Perspective. Factor analysis confirmed our six-factor hypothesis underlying the 28 items.

No abstract

No abstract

This paper describes a connection admission control (CAC) algorithm for ATM networks supporting different Quality of Service (QoS) classes, and illustrates its effectiveness with simulation results considering both the call-level and the cell-level dynamics. The CAC algorithm groups connection requests in three different QoS classes: i) Class 1: with stringent CLR (Cell Loss Ratio) and CDV (Cell Delay Variation) requirements; ii) Class 2: with stringent CLR requirements, but no need for CDV guarantees; iii) Class U: with no need for guarantees on either CLR or CDV. Both Constant Bit Rate (CBR) and Variable Bit Rate (VBR) connections can request admission as either Class 1 or Class 2, depending on their QoS requirements. Unspecified Bit Rate (UBR) and Available Bit Rate (ABR) connections instead normally request admission as Class U. The investigation of the effectiveness of the CAC algorithm is based on the simulation of an ATM network with parking lot topology, and with variable parameter values. The relationship between the proposed CAC algorithm and equivalent bandwidth (EB) CAC algorithms described in the literature is discussed.