Although the infographic and design communities have used simple pictographic representations for decades, it is still unclear whether they can make visualizations more effective. Using charts, we tested how impact (1) memory information just viewed, as well under load of additional information, (2) speed finding (3) engagement preference in seeking out these visualizations. We find that superfluous images distract. But no user costs -- some intriguing benefits when pictographs are to represent data.

Many species, including humans, display group behavior. Thus, perceiving crowds may be important for social interaction and survival. Here, we provide the first evidence that humans use ensemble-coding mechanisms to perceive behavior of a crowd people with surprisingly high sensitivity. Observers estimated headings briefly presented point-light walkers differed in number their members (i.e., differently sized had identical or increasingly variable directions walking). We found observers...

Ensemble coding supports rapid extraction of visual statistics about distributed information. Researchers typically study this ability with the goal drawing conclusions how such extracts information from natural scenes. Here we argue that a second domain can serve as another strong inspiration for understanding ensemble coding: graphs, maps, and other presentations data. Data visualizations allow observers to leverage their perform on distributions spatial or featural estimate actual We...

In this paper, we explore how the capacity limits of attention influence effectiveness information visualizations. We conducted a series experiments to test visual feature type (color vs. motion), layout, and variety elements impacted user performance. The tested users' abilities (1) determine if specified target is on screen, (2) detect an odd-ball, deviant target, different from other visible objects, (3) gain qualitative overview by judging number unique categories screen. Our results...

Several fields of science are experiencing a "replication crisis" that has negatively impacted their credibility. Assessing the validity contribution via replicability its experimental evidence and reproducibility analyses requires access to relevant study materials, data, code. Failing share them limits ability scrutinize or build-upon research, ultimately hindering scientific progress.

The connected scatterplot visualizes two related time series in a and connects the points with line temporal sequence. News media are increasingly using this technique to present data under intuition that it is understandable engaging. To explore these intuitions, we (1) describe how paired relationships appear scatterplot, (2) qualitatively evaluate well people understand trends depicted format, (3) quantitatively measure types frequency of misinter pretations, (4) empirically whether...

Transparent statistics is a philosophy of statistical reporting whose purpose scientific advancement rather than persuasion. We propose SIG to discuss problems and limitations in practices HCI options for moving the field towards clearer more reliable ways writing about experiments.

Transparent statistics is a philosophy of statistical reporting whose purpose scientific advancement rather than persuasion. We ran SIG at CHI 2016 to discuss problems and limitations in practices HCI options for moving the field towards clearer more reliable ways writing about experiments, received an overwhelming response. This resulted rough drafts reviewer guidelines, resources authors, other suggestions advancing vision transparent within field; this year, we propose concentrated...

Replications are rare in visualization research, but if they were more common, it is not unreasonable to believe that would show a similar rate of unreproducible results as the psychological and social sciences. While replication crisis research be helpful wake-up call, examining correcting underlying problems many studies ultimately productive.In this paper, we survey state research. We examine six threats validity suggest ways address them. Finally, describe possible models for publishing...

Two fundamental tenants of scientific research are that it can be scrutinized and built-upon. Both require the collected data supporting materials shared, so others examine, reuse, extend them. Assessing accessibility these components paper itself serve as a proxy for reliability, replicability, applicability field's research. In this paper, I describe current state openness in visualization provide suggestions authors, reviewers, editors to improve open practices field. A free copy data,...

Though the mediums for visualization are limited, potential dimensions of a dataset not. In many areas scientific study, understanding correlations between those and their uncertainties is pivotal to mining useful information from dataset. Obtaining this insight can necessitate visualizing relationships among temporal, spatial, other dimensionalities data its uncertainties. We utilize multiple views interactive exploration selection important features, we apply techniques unique challenges...

Transparent statistics is a philosophy of statistical reporting whose purpose scientific advancement rather than persuasion. At our CHI 2017 workshop, "Moving Statistics Forward", we identified that an important first step to develop detailed guidelines for authors and reviewers in order help them practice promote transparent statistics. We propose SIG solicit feedback from the community on working draft "Transparent Guidelines" engage potential contributors push movement forward.

Preregistration can force researchers to front-load a lot of decision-making an early stage project. Choosing which preregistration platform use must be therefore one those decisions, and because cannot moved, that choice is permanent. This article aims help who are already interested in choose by clarifying differences between them. criteria features explained analyzed for sites cater broad range research fields, including: GitHub, AsPredicted, Zenodo, the Open Science Framework (OSF),...

Replications are rare in visualization research, but if they were more common, it is not unreasonable to believe that would show a similar rate of unreproducible results as the psychological and social sciences. While replication crisis research be helpful wake-up call, examining correcting underlying problems many studies ultimately productive.In this paper, we survey state research. We examine six threats validity suggest ways address them. Finally, describe possible models for publishing...

Several fields of science are experiencing a "replication crisis" that has negatively impacted their credibility. Assessing the validity contribution via replicability its experimental evidence and reproducibility analyses requires access to relevant study materials, data, code. Failing share them limits ability scrutinize or build-upon research, ultimately hindering scientific progress.Understanding how diverse research artifacts in HCI impact sharing can help produce informed...

This paper demonstrates the prevalence of a shared characteristic between visualizations and images nature. We have analyzed visualization competitions user studies found that more preferred, better performing exhibit natural characteristics. Due to our brain being wired perceive [SO01], testing for properties similar those can help show how well is capable absorbing data. In turn, metric finds visualization's similarity image may determine effectiveness visualization. results comparing...

Visualizing the time-variant results of a simulation can help scientists see patterns that would be difficult to find using only statistics. Furthermore, visualizing differences between multiple simulations let directly analyze simulations' consistency. In this article, we describe our interactive application for viewing cosmological simulations. Using programmable shaders and visualization techniques, illustrates how properties millions particles vary across doing so, aim in variance...

Optic flow is the coherent motion of a region in visual field, and system astoundingly effective at perceiving this large-scale complex percept. Whenever you rotate your head or move through scene, nearly every part field moves different speed velocity. For example, as walk forest, trees horizon appear to slowly expand, on sides rapidly extremities vision disappear. In spite such myriad local motions, all information concisely summarized moving forward. Such summary encoding large variety...

Graphs and data visualizations can give us a visual sense of trends on topics ranging from poverty, the spread diseases, popularity products, etc. What makes graphs useful is our ability to perceive these at-a-glance. Related work has investigated effect different properties graphs, including axis scaling, choice encoding, presence pictographic elements (e.g., Haroz et al. 2015) perception or remembered size quantities depicted. Previous shown that attention directed towards text...

When perceiving groups of features, we may rely on ensemble codes pooled feature signals that precisely describe the average (i.e., gist) but limit access to details (Haberman & Whitney, 2007). Here demonstrate use an code perceive direction a walking crowd people. We briefly presented crowds twelve non-overlapping point-light walkers containing different amounts variability had identical or increasingly divergent headings). Observers then estimated crowd. used equivalent noise model (Dakin,...

Across science, education, and business, we process communicate data visually. One bedrock finding in visualization research is a hierarchy of precision for perceptual encodings data, e.g., that encoding with Cartesian positions allows more precise comparisons than sizes. But his has only been tested single value comparisons, under the assumption those lessons would extrapolate to multi-value comparisons. We show when comparing averages across multiple points, even pairs these differences...

Those of us who study large effects may believe ourselves to be unaffected by the reproducibility problems that plague other areas. However, we will argue initiatives address crisis, such as preregistration and data sharing, are worth adopting even under optimistic scenarios high rates replication success. We searched text articles published in Journal Vision from January through October 2018 for URLs (our code is here: https://osf.io/cv6ed/) examined them raw data, experiment code, analysis...