An official website of the United States government
People often search for information in order to learn something new. In recent years, the “search-as-learning” movement has argued that search systems should be better designed to support learning. Current search systems (especially Web search engines) are largely designed and optimized to fulfill simple look-up tasks (e.g., navigational or fact-finding search tasks). However, they provide less support for searchers working on complex tasks that involve learning. Search-as-learning studies have investigated a wide range of research questions. For example, studies have aimed to better understand how characteristics of the individual searcher, the type of search task, and interactive features provided by the system can influence learning outcomes. Learning assessment is a key component in search-as-learning studies. Assessment materials are used to both gauge prior knowledge and measure learning during or after one or more search sessions. In this paper, we provide a systematic review of different types of assessments used in search-as-learning studies to date. The paper makes the following three contributions. First, we review different types of assessments used and discuss their potential benefits and drawbacks. Second, we review assessments used outside of search-as-learning, which may provide insights and opportunities for future research. Third, we provide recommendations for future research. Importantly, we argue that future studies should clearly define learning objectives and develop assessment materials that reliably capture the intended type of learning. For example, assessment materials should test a participant’s ability to engage with specific cognitive processes, which may range from simple (e.g., memorization) to more complex (e.g., critical and creative thinking). Additionally, we argue that future studies should consider two dimensions that are understudied in search-as-learning: long-term retention (i.e., being able to use what was learned in the long term) and transfer of learning (i.e., being able to use what was learned in a novel context).
more »
« less
Cheyenne Demonstratives: A Corpus Study
This paper investigates the eight demonstratives found in Cheyenne, an Algonquian language spoken in Montana and Oklahoma. Demonstratives, such as 'this' and 'that' in English, are used to indicate an individual or object being referred to, either something in the speech context (deictic use) or something that was previously mentioned (anaphoric use). In this paper, we discuss ±proximal and ±animate distinctions made amongst the Cheyenne demonstratives, their frequency of use, and the ways they are used, based on a corpus study of 62 texts (Leman 1980b) and the dictionary (Fisher et al. 2006).
more »
« less
- Award ID(s):
- 1664462
- PAR ID:
- 10487714
- Publisher / Repository:
- Michigan State University Press
- Date Published:
- Journal Name:
- Papers of the Fifty-Second Algonquian Conference
- ISBN:
- 9781611864465
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
null (Ed.)Training competitive deep video models is an order of magnitude slower than training their counterpart image models. Slow training causes long research cycles, which hinders progress in video understanding research. Following standard practice for training image models, video model training has used a fixed mini-batch shape: a specific number of clips, frames, and spatial size. However, what is the optimal shape? High resolution models perform well, but train slowly. Low resolution models train faster, but are less accurate. Inspired by multigrid methods in numerical optimization, we propose to use variable mini-batch shapes with different spatial-temporal resolutions that are varied according to a schedule. The different shapes arise from resampling the training data on multiple sampling grids. Training is accelerated by scaling up the mini-batch size and learning rate when shrinking the other dimensions. We empirically demonstrate a general and robust grid schedule that yields a significant out-of-the-box training speedup without a loss in accuracy for different models (I3D, non-local, SlowFast), datasets (Kinetics, Something-Something, Charades), and training settings (with and without pre-training, 128 GPUs or 1 GPU). As an illustrative example, the proposed multigrid method trains a ResNet-50 SlowFast network 4.5 x faster (wall-clock time, same hardware) while also improving accuracy (+ 0.8% absolute) on Kinetics-400 compared to baseline training. Code is available online.more » « less
-
Human speakers have an extensive toolkit of ways to express themselves. In this paper, we engage with an idea largely absent from discussions of meaning in natural language understanding—namely, that the way something is expressed reflects different ways of conceptualizing or construing the information being conveyed. We first define this phenomenon more precisely, drawing on considerable prior work in theoretical cognitive semantics and psycholinguistics. We then survey some dimensions of construed meaning and show how insights from construal could inform theoretical and practical work in NLP.more » « less
-
null (Ed.)This paper outlines a new model of language revitalisation that understands language to be a characteristic of a nexus of social activities rather than an independent object. Language use is one of an overall set of factors contributing to the wellbeing of a particular community. Our model treats language as one node (or a cluster of nodes) in a complex system of interacting behaviours. Changes to another node or in the language node(s) itself can impact overall social wellbeing, something often ignored by linguists (but not by other social scientists working in Indigenous communities). Disruption to an existing network occurs within a time frame; the longer the disruption, the more likely that the network redefines the group. Variables that define the language ecology operate on multiple levels. For the group and for individuals within the group, there can be considerable variation in usage and proficiency over time. Sustainability cannot be reduced to simple cause-and-effect relationships between sociocultural variables. The next phase of language revitalisation projects should be built around the concept of language activity as part of promoting community wellbeing. The use of complex networks that have been applied to human wellbeing in other contexts support our argument.more » « less
-
Abstract The phenomenon of ambiguity aversion suggests that people prefer options that offer precisely rather than imprecisely known chances of success. However, past work on people's responses to ambiguity in health treatment contexts found ambiguity seeking rather than aversion. The present work addressed whether such findings reflected a broad tendency for ambiguity seeking in health treatment contexts or whether specific attributions for ambiguity play a substantial role. In three studies, people choose between two treatment options that involved similar underlying probabilities, except that the probabilities for one option involved ambiguity. The attributions offered for the ambiguity played an important role in the results. For example, when the range of probabilities associated with an ambiguous treatment was attributed to the fact that different studies yield different results, participants tended to show ambiguity aversion or indifference. However, when the range was attributed to something that participants could control (e.g., regular application of a cream) or something about which they were overoptimistic (e.g., their immune system function), participants tended to show ambiguity seeking. Health professionals should be mindful of how people will interpret and use information about ambiguity when choosing among treatments.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Proceeding:
- The DOI is not currently available.
