The Now and Then: Integrating Current and Historical Data in Small Multiple Time Series Visualization
Listen
Radio episode about this paper
Transcript
Introduction to the show: ident: Computer Science. The day in computer science: software, systems and theory on arXiv.
Tom: I'm Tom.
Jane: And I'm Jane, here with Lu, Meng and our in-house large language model, Lalam.
Tom: We're looking at the paper: "The Now and Then".
Title and authors: Tom: So we're looking at this paper called The Now and Then: Integrating Current and Historical Data in Small Multiple Time Series Visualization, and it’s about making those little charts that show a lot of data at once—like in healthcare or manufacturing—better.
Jane: That’s right, Tom. It tackles the problem that people monitoring a bunch of things can get overwhelmed because they need to see what's happening right now while also watching how things have been trending over time.
Lu: Basically, the paper is saying that we don't have good design rules yet for visually highlighting those current values in historical charts represented by small multiples.
Meng: I mean, if you’re looking at a grid of hundreds of little time series, you need a way to quickly spot if something is spiking or dropping without losing track of the bigger picture.
Lalam: It sets up a design space for glanceable displays that balances showing what's happening now with keeping the history in context.
The paper's summary: Tom: The authors looked at this problem by building a design space to figure out how to integrate current and historical values without messing up the trend reading.
Jane: They tested four different ways of doing this: Standard, Emphasis, Integrated, and Separated. They wanted to see which way worked best for rapidly checking current values while still letting you follow the trend.
Lu: The key finding they highlighted is that integrated designs were better than separated ones for tasks where people had to look up specific information.
Meng: So they found that putting the current value inside the line chart, instead of just next to it, made it easier for users to appraise what was happening instantly while still seeing the time series context.
Lalam: They also looked at how size and color could be used for emphasis when you’re dealing with specific thresholds or ranges.
The paper's improvements: Tom: One of the big takeaways they found is that integrated designs had less interference than separated ones when users were trying to find current values.
Jane: That means users can rapidly check those current values without having to hunt around too much, which is huge in high-stakes environments.
Lu: They also looked at size encoding for the current value when people are performing non-threshold appraisal tasks, and they found that putting the size encoding inside the line chart was a good compromise.
Meng: That means if you're trying to find a very large or very small current value, making it bigger on the screen helps without ruining your view of how the data has been behaving before.
Lalam: And for when you have data points that cross certain limits, they found that color threshold encodings worked much better than using shaded bands to show those thresholds in dense settings.
Conclusion: Tom: So, overall, the paper "The Now and Then: Integrating Current and Historical Data in Small Multiple Time Series Visualization" suggests that integrated designs with size encoding for current values and color threshold encodings are the most effective ways to help people monitor data at a glance.
Jane: It’s about giving users a way to spot immediate problems without having to switch back and forth constantly between the current reading and the historical trend.
Lu: The main implication is that we can design better interfaces for these complex monitoring scenarios by focusing on how visual channels interact with the data structure itself.
Meng: From an engineering standpoint, this means when we build these systems, we should prioritize that integrated approach to keep things fast and clear under pressure.
Lalam: I’m pretty confident in this direction because the experiments showed a measurable improvement in how quickly people could appraise those current values compared to just looking at the trends alone.
Northeastern University
cs.HC
Submitted: 2026-10-06
Updated: 2026-10-06
Comments: 16 pages, 9 figures, to be published in IEEE TVCG
DOI: 10.1109/TVCG.2026.3740740
License: http://creativecommons.org/licenses/by/4.0/
Importance score: 71/100
The gist: Small multiple visualizations are often used for real-time data monitoring tasks in high-impact domains such as healthcare and manufacturing Effective design is critical because users rely on these…
Key concepts
- Small Multiple Visualizations
- These are multiple small charts displayed together to monitor many data entities at once. They are common in high-impact fields like healthcare or manufacturing, but designers lack clear guidelines on how to emphasize the most recent data while keeping historical trends readable.
- Design Space Categories
- The paper proposes four ways to combine current and historical data: Standard, Emphasis, Integrated, and Separated. The goal is to find the best balance between showing what is happening now and understanding the past without confusing the user.
- Integrated Current Value Designs
- This design approach places the current value directly onto or within a line chart. The study shows this method minimizes interference with reading historical trends while allowing users to rapidly assess current data points, which is crucial for glanceable monitoring.
Terminology
Summary
Small multiple visualizations are often used for real-time data monitoring tasks in high-impact domains such as healthcare and manufacturing Effective design is critical because users rely on these visualizations to monitor data from many entities, such as patients or machines, often while distracted No design guidelines currently exist for visually emphasizing current values in historical time series represented by small multiples. p. 0 Purdue, Enrico Bertini, and Melanie Tory
Abstract—Small multiple time series visualizations are often
used for real-time data monitoring tasks in high-impact domains
such as healthcare and manufacturing. Effective design is critical
because users rely on these visualizations to monitor data
from many entities, such as patients or machines, often while
distracted. However, no design
guidelines currently exist for visually emphasizing current values
in historical time series represented by small multiples. Via an
iterative design process informed by a review of related literature
and theory on visual channels and emphasis, we present a
design space for glanceable time series small multiple displays.Where the paper says it
Where the paper says it
Design Space Characterization
The authors developed a design space for integrating current and historical values in time series small multiple visualizations to explore how to emphasize current values without interfering with trend reading. This space consists of four categories: Standard, Emphasis, Integrated, and Separated. p. 3 Design Space
Our design space has four categories addressing the relationship between the representation of the current value and
the historical trend present in time series small multiples. The categories are Standard, Emphasis, Integrated, and
Separated.Where the paper says it
Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.
Improvements for AI systems
-
Prioritize integrated designs for glanceable monitoring displays because "Integrated current value designs had lower overall interference than separated ones for the Lookup & Find tasks and design variations that we examined.
This allows users to
rapidly appraise current values" while retaining temporal context, which is critical in high-impact domains like healthcare and manufacturing. -
Use size encoding on the current value when performing non-threshold appraisal tasks because "size encodings on the current value, spatially integrated into the line chart, may provide a good compromise, with 28% response time improvement for tasks involving finding large current values and minimal interference with trend lookup tasks." This optimizes performance for finding high or low values without significantly hindering trend assessment.
-
Employ color threshold encodings instead of shaded band encodings when monitoring threshold-based data because
color threshold encodings significantly outperformed shaded band encodings
in experimental settings, suggesting color is more effective for indicating thresholds in dense scenarios like small multiples.
Abstract
Small multiple time series visualizations are often used for real-time data monitoring tasks in high-impact domains such as healthcare and manufacturing. Effective design is critical because users rely on these visualizations to monitor data from many entities, such as patients or machines, often while distracted. Users may need to rapidly appraise current values for each entity, monitoring for those that go outside an acceptable range, while also watching temporal trends. However, no design guidelines currently exist for visually emphasizing current values in historical time series represented by small multiples. Via an iterative design process informed by a review of related literature and theory on visual channels and emphasis, we present a design space for glanceable time series small multiple displays. We evaluate this space through two online empirical studies, testing against non-threshold and threshold rapid appraisal tasks. Our results provide insights into merging current value and historical data visualizations for rapid appraisal tasks in time series monitoring. For non-threshold tasks, we found that size encodings on the current value, spatially integrated into the line chart, may provide a good compromise, with 28% response time improvement for tasks involving finding large current values and minimal interference with trend lookup tasks. More generally, integrated designs outperformed separated designs (in which the current value representation is spatially separated from the historical trend line). For threshold tasks, color threshold encodings significantly outperformed shaded band encodings. All supplemental materials are available at https://osf.io/wzf6a.
Related papers
- EmoPhone: A Multi-Wave Dataset for In-the-Wild Mobile and Wearable Affect Sensing
- Tasteply: Externalizing Stable and Situational Preference as Manipulable Spatial Objects
- Shielded Trajectory Sampling for Personalized Relational Safety in Conversational AI
- Reconstruction and Reflection of Positive Experiences through Resurfacing Laughter-indexed Everyday Moments
- LLM-Assisted Reranking to Operationalize Nuanced Objectives in Recommender Systems
- Explainable Iterative Data Visualization Refinement via an LLM Agent