428The Internet has revolutionized the way in which people communicate and retrieve information. Powerful communication tools are transforming many scientific disciplines, including experimental and clinical psychology. The Web allows access to much wider populations-as well as to populations that were previously difficult to reach-in an inexpensive, fast, and convenient way. In clinical psychology, for example, psychological testing and assessment can be done online (see, e.g., Buchanan, 2002).In experimental psychology, the Web is increasingly being used as an alternative to traditional lab settings for running experiments. In the present article, we review the major advantages and disadvantages of online experiments in comparison with traditional lab experiments; then we compare online versus lab results in problem solving-an area in which this comparison has received less attention. Online Experiment Pros and ConsThere are many potential advantages for doing an experiment online as opposed to in the lab (see Birnbaum, 2004, for a review of pros and cons). First, experimental procedures can be automated, thus reducing costs and the amount of time spent managing the experiment (Reips, 2002a). This also increases the uniformity of the procedure across participants and may reduce demand characteristics (Reips, 2002a). Second, online experiments can be done in a wider array of settings-not just in the highly constrained setting of the lab (Reips, 2000)-and can include 24-h access (Reips, 2002a), considerations that can increase participants' comfort (Salgado & Moscoso, 2003). Third, ethical standards can be maintained because the experiment is publicly available for criticism and the possibility for the coercion of participants is reduced (Reips, 2002a). Finally, online accessibility allows the targeting of specific audiences (through mailing lists or newsgroups) and broadens the participant pool to Web users, rather than, for example, undergraduate students at a particular university, which may allow increased generalizability of the results (Reips, 2000).There are also disadvantages to running an experiment online rather than in the lab. First, the environments will be more variable, including noise, lighting, and technical aspects of the equipment. Effects of this variability may be reduced by asking participants to do the study in a particular sort of environment and by checking for statistical outliers. Second, online experiments are vulnerable to multiple submissions. This seems to be generally rare (Reips, 2000), but it may be more likely when participants have strong opinions about the topic (see, e.g., Konstan, Rosser, Ross, Stanton, & Edwards, 2005). The risk of multiple submissions can be reduced by asking for personal information, using password protection or an IP address verification (Reips, 2002b), and by reducing external incentives, such as winning money or a prize. Finally, there may be biases in the final sample: Only interested and motivated participants may start (self-selection) and complete...
WOS:000274672400002International audienceNeural networks were trained with backpropagation to map location-specific letter identities (letters coded as a function of their position in a horizontal array) onto location-invariant lexical representations. Networks were trained on a corpus of 1179 real words, and on artificial lexica in which the importance of letter order was systematically manipulated. Networks were tested with two benchmark phenomena - transposed-letter priming and relative-position priming - thought to reflect flexible orthographic processing in skilled readers. Networks were shown to exhibit the desired priming effects, and the sizes of the effects were shown to depend on the relative importance of letter order information for performing location-invariant mapping. Presenting words at different locations was found to be critical for building flexible orthographic representations in these networks, since this flexibility was absent when stimulus location did not vary
We studied the feedforward network proposed by Dandurand et al. (2010), which maps location-specific letter inputs to location-invariant word outputs, probing the hidden layer to determine the nature of the code. Hidden patterns for words were densely distributed, and K-means clustering on single letter patterns produced evidence that the network had formed semi-location-invariant letter representations during training. The possible confound with superseding bigram representations was ruled out, and linear regressions showed that any word pattern was well approximated by a linear combination of its constituent letter patterns. Emulating this code using overlapping holographic representations (Plate, 1995) uncovered a surprisingly acute and useful correspondence with the network, stemming from a broken symmetry in the connection weight matrix and related to the group-invariance theorem (Minsky & Papert, 1969). These results also explain how the network can reproduce relative and transposition priming effects found in humans.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.