In this dissertation, I explore determinants, and some consequences, of persistence in consumer search. Many prominent thinkers have considered the problem of search in terms of optimal solutions, or their heuristic approximations. In the following research, I explore persistence in search not merely as a function of economic calibration, but rather as an outcome determined by both cognitive and motivational processes. I provide evidence that normative models of search are insufficient to explain the behavior of those whom I study. Instead, I show cases in which search persistence is a function of prior behavior (Chapter 1) and prior beliefs (Chapter 2). I further propose a cognitive model of price search behavior (Chapter 3) that can predict many of the observed behaviors that would be considered mistakes in normative price search frameworks (e.g., variance neglect, reference point effects, local contrast effects).

This thesis is concerned with the psychology of magic and illusion. In particular it is focused on three illusions, which can be conceptualized as types of invisibility: (1) Illusions of omission - failures to see, (2) Illusions of commission - seeing things that are not really present, and (3) metacognitive illusions - people's false beliefs about their own cognitive and perceptual systems. The work presented in this thesis is set out to explore these illusions through behavioural visual experiments inspired by sleight-of-hand magic tricks. Across three distinct paradigms using stimuli ranging in complexity from static line drawings, to recorded videos, to live events, I demonstrate that manipulations of observers' expectations and attentional set can result in perceptions of visual events that are variously accurate representations, illusions of omission, or illusions of commission. I also demonstrate that these illusions are often associated with failures of visual metacognition, in that they are generally considered to be surprising and counterintuitive. In addition to these empirical elements of the project, I also consider historical and contemporary connections between experimental psychology and magic tricks. I show that, in some instances, magicians' misdirection techniques anticipated developments by experimental psychology by hundreds of years, and that the idea of investigating the mechanisms of magic tricks and illusions played a key role in the development of experimental psychology as a scientific discipline. Through this combination of historical analyses novel experiments, I show that the integration of magic and experimental psychology has a great potential to drive future research in human cognition and perception.

A comparison of two processes for the control of mentation was undertaken. I compared externally mandated thought suppression to more internally motivated mood management to clarify the mechanisms whereby insulted men regulate and repair their angry mood state. The one hundred and eight male undergraduates completed the Beck-Depression Inventory and the Vengeance Scale, then volunteered for a one-hour study about 'Education and the Job Market.' The men were assigned randomly to one of six treatment conditions of a 2 (Suppression Relevance: suppression of an anger-relevant word vs. suppression of an anger-irrelevant word) x 3 (Retaliatory Motivation: insult-retaliation vs. insult-no retaliation vs. no insult-no retaliation) factorial design; Stimulus Word Valence (affectively angry stimulus words vs. affectively neutral stimulus words) served as a within-subject variable I predicted that when insulted men not anticipating retaliation were told to suppress an anger-relevant word, they would generate word associations higher in anger-related content and respond faster than their baseline controls. Marginally supportive findings indicated that this enhanced accessibility of anger-related word associations occurred when the men gave their second response to angry stimulus word prompts, but not when responding to neutral ones. Methodological limitations and improvements are offered, and findings are discussed by comparing the thought suppression paradigm of Wegner and Erber (1992; Wegner, Erber, and Zanakos, 1993) to the mood management model of Zillmann and Bryant (1985; Taylor, 1992) / acase@tulane.edu

The question of how to detect multivariate outliers has presented both philosophical and statistical problems. The method most widely used for the detection of multivariate outliers is Mahalanobis' D-Squared statistic (D$\sp2$), commonly viewed as analogous to a univariate standard score. D$\sp2$ is simple to calculate, and its asymptotic distribution is known to be the Chi-square distribution (Barnett, 1976, 1978a, 1978b, 1979; Beckman & Cook, 1983; Hawkins, 1974, 1980). Additionally, D$\sp2$ or D$\sp2$/df is available from major statistical packages such as BMDP and SPSSX. The distribution of D$\sp2$s calculated using the sample centroid and variance-covariance matrix is thought to be mathematically intractable (Barnett, 1984; Wilks, 1963). Some researchers have suggested the use of ordered Chi-square (Barnett, 1984; Beckman & Cook, 1983; Hawkins, 1980) or ordinary Chi-square (Comrey, 1984; Rasmussen, 1988; Tabachnick & Fidell, 1983) critical values for evalaution of D$\sp2$ in the detection of outliers. Tables of ordered Chi-square critical values are not available, thus, it was necessary to compute these values for the present study. This study examined the fit of D$\sp2$ with the Chi-square and ordered Chi-square distributions, via Monte Carlo Methods, and determined that neither provided accurate critical values. Consequently, critical values were generated empirically. The resulting tables of critical values cover the largest 25% of ordered D$\sp2$ for the conditions resulting from a full factorial cross of numbers of subjects (20, 30, 40, 50, 100, 200, 300, 500, 1000), and numbers of variables (2, 3, 4, 5, 6, 7, 8, 9, 10, 15, 20, 25, 30, 35, 50) / acase@tulane.edu

Manipulations of central cholinergic activity alter the female rodent sexual behavior, lordosis. Agents that enhance cholinergic function, such as the acetylcholinesterase inhibitor physostigmine, activate lordosis in intact cycling female rats during proestrus when endogenous estrogen levels are high and not during diestrus when endogenous estrogen levels are low. In the present experiments, intraventricular administration of physostigmine activated lordosis in intact female rats at Mid-diestrus or Diestrus II only when exogenous estrogen was administered the day before. The degree of activation of lordosis by physostigmine at Mid-diestrus or Diestrus II was related to the priming dose of estrogen. These results indicate that cholinergic mechanisms that regulate sexual receptivity depend upon sufficient estrogen priming. Serum progesterone titers measured in these females were higher at Mid-diestrus than Diestrus II. In the absence of cholinergic manipulations, females were more likely to exhibit lordosis at Mid-diestrus than at Diestrus II, possibly as a consequence of elevated progesterone levels at Mid-Diestrus. Despite dissimilar levels of circulating progesterone, physostigmine activated lordosis equally at Mid-diestrus and Diestrus II following estradiol priming. These results suggest that the level of progesterone did not influence the ability of the cholinergic system to activate lordosis. Muscarinic binding as determined by the muscarinic antagonist ($\sp3$H) QNB varied across the estrous cycle. When compared to muscarinic binding during diestrus, muscarinic binding during Proestrus-Estrus was increased in two brain regions traditionally associated with the activation of lordosis, the medio-basal hypothalamus and central gray, and decreased in a brain region traditionally associated with the inhibition of lordosis, the preoptic area. However, administration of estradiol on the day before decreased muscarinic binding in the septum, preoptic area, and central gray. Estradiol administration did not change muscarinic binding in the medio-basal hypothalamus. Because these changes in binding were complex and unpredicted, a relationship between the estrogen regulation of lordosis and muscarinic binding in these areas was not evident. These results do not support the hypothesis that changes in muscarinic binding in these brain regions are necessary for the cholinergic activation of lordosis / acase@tulane.edu

The independent effects of luminance and chromatic contrast between text and background on reading speed were examined for normal and reading disabled subjects. The requirement of making eye movements for reading was also manipulated. The factors of chromatic contrast without luminance contrast (equiluminance) and reading without required eye movements (with rapid serial visual presentation, or RSVP) disrupt and minimize, respectively, the need for visual spatial localization of gaze within lines of text. Without luminance contrast, visual perception of spatial position and motion is disrupted to a certain degree. With RSVP, skilled readers can read two and three times faster than with traditional line by line reading. Reading rates were compared across conditions of contrast (high luminance and equiluminance contrast) and of text presentation (RSVP and traditional page format). Adult skilled readers participated in the first study, and groups of average ability and reading disabled children aged thirteen to sixteen participated in the second study. In Study 1, adults read equiluminant contrast text slower than text with a high luminance contrast. The degree of slowed reading under equiluminance was greater when eye movements were required for reading. In Study 2, normal reading was slowed with equiluminance, while disabled reading was not affected by equiluminance. This contrast effect did not reliably distinguish reading ability groups. Also in Study 2, both normal and disabled reading rates increased significantly when eye movements were not required with RSVP text. Normal reading rates were more facilitated by RSVP than disabled reading rates / acase@tulane.edu

The male genitalia of many invertebrate species are often elaborate, more elaborate than what is necessary for the relatively simple act of sperm transfer. These flamboyant morphologies may be the result of sexual selective processes similar to those which produce elaborate secondary sex characteristics such as brilliant plumage, antlers, or large body size The study investigated palp morphology and male reproductive success in Nephila clavipes, a neo-tropical orb-weaving spider. First, males were placed with sexually receptive females and reproductive activity was recorded and correlated with aspects of gross and palpal male morphology. Second, sperm uptake and storage were assessed by sacrificing mated females and counting the number of sperm within their spermathecae by means of a computer imaging system. The number of sperm counted was then correlated with aspects of male copulatory behavior and morphology. Third, sperm utilization was evaluated by permitting mated females to oviposit and then counting the percentage of eggs in clutch that eventually hatched. Finally, the effect of conductor breakage was evaluated by assessing the timing of its occurrence and its influence on sperm transfer There was less variation in palp size of male Nephila than in other aspects of male morphology. The distribution of palp size was essentially normal, indicating stabilizing selection. Palp size and gross male size were positively correlated. Gross male body size was the best predictor of how much sperm was produced, transferred to, and stored by the female. Size of the male was not related, however, to the percentage of sperm actually transferred. Having smaller palps did not result in any obvious compensatory behavioral strategies. Time of mating influenced how many sperm were retained by the female but there appeared to be little relationship between copulatory behavior itself and the amount of sperm later found. None of the variables analyzed in this study greatly influenced the percentage of eggs eventually hatching. Conductor breakage seriously interfered with sperm transfer but occurred less often than expected and did not appear to be a result of copulatory activity. These findings suggest that, within the parameters of this study, intersexual selection through female choice is not a significant process in this species / acase@tulane.edu

This study investigated the client's perception of the therapist using three types of clinical humor: appropriate, inappropriate, and no humor. These humor conditions were interspersed within identical scripts, which were enacted and videotaped The hypotheses were that: (1) the therapist using appropriate humor would elicit the most positive response, (2) the male subjects would more appreciate clinical humor use, and (3) the male therapist using appropriate humor would be rated the most positively One hundred and twenty subjects used the Counselor Rating Scale by Barak and LaCrosse, to evaluate the video therapist. All subjects had previous therapy experience Appropriate humor did not elicit a more positive score than the use of no humor, but was significantly more positive than inappropriate humor. Male subjects favored inappropriate humor significantly. Further, an interactional trend suggested a gender bias; the male therapist was rated highest when using appropriate humor, whereas the female therapist was rated highest when not using humor / acase@tulane.edu

In two studies, motivational factors which could influence the extent to which negative affect selectively facilitates memory for negative material were investigated. In the first experiment, male undergraduates memorized words with positive, negative, and neutral connotations. The men were then exposed to treatments intended to make them either happy, angry, or unchanged in affect, and either anticipated or did not anticipate a later opportunity to evaluate the individual responsible for their treatment. They were next asked to recall then recognize the words. Angered men in the evaluation (retaliation) condition were found to recall more words with a negative semantic content than other words, and more than all other subjects. In contrast, happy men recalled more positive words regardless of evaluation condition. No recognition differences were found. In the second experiment, memory for details of a provoking experience was tested. Male undergraduates either received or did not receive information that they would later have an opportunity to retaliate against a provocateur, with the information being given either before or after they were provoked. Men informed that they would be able to retaliate more accurately recalled the details of the provoking experience, but the timing of retaliation information had no bearing on memory for these details. Results are discussed in terms of motivation-based selective generation of items in memory / acase@tulane.edu

The neurotoxin kainate produces hippocampal lesions which in turn induce the mossy fibers, the axons of the dentate granule cells, to sprout collaterals. These collaterals grow across the granule cell body layer and appear to synapse on the dendrites of the granule cells. If these synapses are functional, a recurrent excitatory circuit will have formed in the dentate. As hippocampal tissue taken from human temporal lobe epileptics show similar pyramidal cell loss and mossy fiber sprouting, the KA-treated animal has been proposed as a model for temporal lobe epilepsy (Nadler, 1981) Tauck and Nadler (1985) have presented data suggesting mossy fiber sprouting is associated with the generation of multiple population spikes in the dentate in response to hilar stimulation. Because normally only a single population spike is evoked, these data suggest the sprouted mossy fibers have formed a functional circuit. To further assess the physiology of dentate granule cells after mossy fiber sprouting, intracellular and extracellular recordings from hippocampal slices were performed As a preliminary step in examining mossy fiber sprouting in the kainate treated rat, we studied the electrically evoked responses in the dentate gyrus of normal rats under conditions of heightened excitability. After blocking GABA$\sb{\rm A}$ receptors with bicuculline (30 $\mu$M), a second population spike was evoked to hilar stimulation. In higher concentrations of GABA$\sb{\rm A}$ blockers (100 $\mu$M bicuculline or picrotoxin), three to four population spikes were evoked to hilar stimulation. After raising (K$\sp+$) from 3 to 10 mM in 30 $\mu$M bicuculline, four to five population spikes were evoked. These data indicate multiple population spikes to hilar stimulation do not require mossy fiber sprouting Experiments conducted on kainate-treated rats revealed that in normal medium hippocampal slices from both animals with and without mossy fiber sprouting had normal responses. When GABA$\sb{\rm A}$ responses were blocked with 30 $\mu$M bicuculline, rats with mossy fiber sprouting showed bursts of population spikes after long and variable delays to low-intensity hilar stimulation. Some slices from hippocampi with mossy fiber sprouting also showed spontaneous population bursts in bicuculline. These responses were not seen in treated animals which did not sprout mossy fiber collaterals. Based on computer simulations and experiments on other brain areas, these responses suggest local recurrent excitatory circuits have formed. These data also argue that the responses of the dentate gyrus with mossy fiber sprouting are normal, and show abnormality only when inhibition is depressed Mossy fiber sprouting has been documented in human tissue taken from temporal lobe epileptics (Houser et al., 1990). We have developed methods for assessing mossy fiber sprouting in tissue taken from humans during surgery for intractable epilepsy / acase@tulane.edu