Quebec’s residential energy use is better explained by demographics than building age, Concordia study shows

The researchers say combining census data with meter readings could help develop more equitable and efficient energy strategies

Concordia University

Facebook

XLinkedInWeChatBlueskyMessageWhatsAppEmail

 

image: 

Masood Shamsaiee

view more 

Credit: Concordia University

Quebec’s winters are notoriously long, cold and energy intensive. And even though Hydro-Québec provides abundant, relatively inexpensive electricity, waste and efficiency remain serious concerns.

A new Concordia study clarifies energy consumption patterns by looking at neighbourhood-level usage in Quebec’s major urban centres and socio-demographic data derived from Canada’s 2021 census, such as employment and income levels, car ownership, average age and household size. The researchers say this method could help utility managers, policy planners and government authorities better understand which demographics consume the most household energy and where usage is at its peak. The data could also be used in devising more effective and equitable energy strategies.

Hydro-Québec provided the researchers with hour-by-hour usage data from residential smart meters for the full four years between 2019 and 2023. The team then broke down that data for Montreal, Trois-Rivières and Quebec City according to the Forward Sortation Areas — Canada Post’s geographical subdivisions used to sort and deliver mail as efficiently as possible.

Using sophisticated models and advanced machine learning tools, the researchers determined which variables drove energy consumption the most across both long-term heating patterns and short-term daily use.

“Energy demand is not just about the building, but about the people who reside in it,” says lead author Masood Shamsaiee, a PhD student at the Next-Generation Cities Institute. “You can have two neighbourhoods with similar building profiles, but if two different types of people live in them, they are going to have two completely different consumption patterns.”

The study was published in the journal Energy and Buildings. It was co-authored by Ursula Eicker, a professor in the Department of Building, Civil and Environmental Engineering.

Age, income, employment as key drivers

To observe long-term patterns, the researchers used a method called change-point analysis to determine when heating systems were activated as outdoor temperatures changed. They then applied a machine-learning model to measure how different socio-demographic factors influenced heating behaviours.

For short-term patterns, the team grouped daily electricity-use profiles into clusters and used another machine-learning model to determine which social characteristics best predicted each pattern. AI analyses were used so the models could reveal not just predictions, but also which factors mattered most.

The analysis revealed strong links between energy use and social characteristics. Higher-income neighbourhoods and areas with larger households tended to have higher baseline electricity use and stronger increases in consumption as heating demand rose. Lower-income neighbourhoods, however, often activated heating earlier in the season, which may reflect less efficient buildings due to poorer insulation and older windows.

Neighbourhoods with older populations tended to have higher electricity use per person, likely because residents spend more time at home and prioritize indoor comfort. Areas with a higher population of non-Canadians, newer homes, high-rise apartments, younger residents and crowded living conditions averaged lower use.

Daily routines also played a role in consumption demands. Areas with high employment levels and car-dependent lifestyles showed strong peaks in the morning and evening, when residents left for and returned from work. Neighbourhoods with higher unemployment or more walkable environments tended to have flatter electricity-use patterns throughout the day.

“I hope this study adds a human element to the different models informing policy makers,” says Shamsaiee. “This can be used as a tool to help utility companies like Hydro-Québec develop better, more detailed efficiency programs while at the same time delivering a fairer and more equitable distribution system.”

Read the cited paper: “Socio-Demographic insights on urban building energy consumption”

---

Journal

Energy and Buildings

DOI

10.1016/j.enbuild.2025.116793 

Article Title

Socio-Demographic insights on urban building energy consumption

Imagination is more than sensory replay

Higher‑level brain systems that interpret perception may play a central role in imagination, study found

Northwestern University

FacebookXLinkedIn

WeChat
WhatsAppEmail

 

video: 

B-roll of the study author and MRI scanner at Northwestern University. 

view more 

Credit: Kristin Samuelson, Northwestern University

• Study suggests long-standing sensory reinstatement theory needs to be refined
• In precision fMRI scans, vividness of imagination was related to activity in higher-order brain networks
• Findings suggest mental imagery is closely tied to higher-level cognitive functions, as opposed to being a strictly sensory phenomenon

CHICAGO --- Imagination is one of the most powerful things our brains can do. We can relive past events while taking a walk, rehearse future conversations through inner speech or sense the heat of a fire without touching it — allowing us to learn, plan and avoid danger without direct experience.

Why imagination is often accompanied by mental imagery remains a longstanding question. When one thinks of an apple, for example, many “see” an image of an apple in their mind. When one thinks of their favorite song, many “hear” that song playing in their mind, including vocals and specific lyrics. Mental imagery has often been thought to rely mainly on reactivating the brain’s sensory regions in the absence of input — a process known as sensory reinstatement. 

But a new Northwestern University study suggests that higher‑level brain systems that interpret and organize perception may also play a central role in imagination.

The scientists asked study participants to imagine different scenarios, such as a child’s birthday party or a castle on a hill, while undergoing individual‑level precision fMRI scanning. The findings suggest that imagination is not simply a copy of sensation. Instead, it appears to emerge at later stages of processing, when the brain represents information holistically as scenes, words, events or ideas rather than raw sensory input. 

“When you ask someone to imagine the sound of a kid’s birthday party, they don’t just hear it — they also automatically picture the scene,” said senior author Rodrigo Braga, assistant professor of neurology at Northwestern University Feinberg School of Medicine. “It makes sense that imagination operates in this holistic, higher‑level space, given that we use it to plan, understand and speculate.” 

The study will publish March 31 in Neuron. 

The findings suggest mental imagery is closely tied to higher-level cognitive functions, as opposed to being a strictly sensory phenomenon. 

“Our study doesn’t refute sensory reinstatement theory, but it does suggest we need to refine it,” Braga said. “It’s not just the sensory parts of the brain that are involved. When people imagine rich scenes or an internal dialogue, the strongest overlap with perception appears in later stages, where sensation has already been transformed into meaning.”

How the study worked

Eight study participants imagined different scenarios during eight separate MRI sessions, as part of a study that resulted in more than 60 hours of fMRI data. The scientists mapped each participant’s sensory and association networks and compared brain activity during imagination with activity during actual perception. They found brain activity related to imagining and perception overlapped in the higher-level association areas, not the early sensory areas. 

“These association areas are particularly interesting because they are greatly expanded in the human brain compared to our close evolutionary ancestors,” Braga said. “They also allow humans to do things we are particularly advanced at, such as communicating using language. This suggests that the generation of mental imagery relies on brain networks that are particularly prominent in the human brain and suggests that these association areas likely work with earlier sensory parts of the brain to institute mental experiences.”

After exiting the scanner, participants self-reported what they had imagined in the scanner, which allowed the scientists to relate each participant's subjective reports about each imagined item with their own brain activity patterns. The participants reported that they experienced vivid visual imagery when imagining scenes and vivid sound imagery when imagining speech.

The data supported two insights into mental imagery: First, different types of imagining activated different networks, Braga said. When people thought about scenes, they were tapping into parts of a brain network called the “default network” — which works with the hippocampus (a key memory structure) to support internally generated thought such as when we think about the past or future. But when they used inner speech or thought about speech, they activated a different network called the “language network.”

Despite these differences, in both cases, imagination overlapped with perception primarily in high-level, transmodal brain regions rather than sensory-specific regions.

“The default network has sometimes been implicated as the brain’s ‘hub’ for mental imagery,” said first author Nathan Anderson, a former postdoctoral fellow at Northwestern. “Our results do show that the default network is generally engaged during imagining, but we also see different large-scale brain networks activated depending on what you are imagining.”

Activity in these association regions also tracked how vivid participants reported their imagery to be, suggesting that naturalistic imagination relies especially strongly on higher‑order interpretive systems, Braga said. The results advance our understanding of how the brain supports self-generated and sensory-independent forms of thought, Braga said. 

The findings do not mean the brain’s sensory cortex is irrelevant, Braga emphasized, but they suggest a more nuanced understanding of how the brain generates mental imagery.

The study is titled, “Mental imagery and perception overlap within transmodal association networks.” 

Brain scan during MRI (IMAGE)

Northwestern University

FacebookXLinkedInWeChatBlueskyMessageWhatsAppEmail

 

A scan of a brain during an MRI

Credit

Kristin Samuelson, Northwestern University

---

Journal

Neuron

Article Title

Mental imagery and perception overlap within transmodal association networks

Article Publication Date

31-Mar-2026

 

The safest cabin layout for efficient aircraft evacuation

To evacuate quickly in case of an emergency, elderly passengers, who may be limited in dexterity, should be evenly distributed among aircraft cabins.

American Institute of Physics

LinkedInWeChatBlueskyMessageWhatsAppEmail

 

image: 

A visualization of the constructed cabin model the researchers used to simulate evacuation scenarios.

view more 

Credit: Zhao et al.

WASHINGTON, March 31, 2026 — In case of an emergency, the Federal Aviation Administration requires aircraft to be able to evacuate within 90 seconds. However, as the median age of the global population increases, the growing number of elderly airline passengers poses new challenges during emergency situations.

In AIP Advances, by AIP Publishing, an international collaboration of researchers simulated 27 different evacuation scenarios in case of a dual-engine fire in an Airbus A320, one of the most common narrow-body aircraft in the world. They compared three different cabin layouts with three different ratios of passengers over the age of 60 and three different distributions of those passengers.

“While a dual-engine fire scenario is statistically rare, it falls under the broader category of dual-engine failures and critical emergencies in aviation. History has shown that dual-engine failures and emergencies, such as the famous ‘Miracle on the Hudson’ involving Captain Sullenberger, can happen and lead to severe consequences,” said author Chenyang (Luca) Zhang. “Our study focuses on these low-probability but high-impact events to ensure the highest safety standards.”

In seeking the most efficient combination of factors, the researchers created full-scale computer-aided design models of the A320 cabin and used Pathfinder — the industry-standard software for evacuation modeling — to simulate passengers’ behavior. They found the proportion and location of elderly passengers have the largest effect on evacuation time.

The fastest option — a layout that accommodates a total of 152 passengers with two rows of first-class seats at the front, and 30 elderly passengers evenly distributed throughout the cabin — still required 141 seconds for all the passengers to reach the ground, much longer than the FAA mandates.

Previous studies have shown that cognitive decline in elderly populations can affect situational awareness and delay decision making, and that reduced dexterity can be exacerbated during high-stress situations. The researchers hope that incorporating this information into their findings — for example, by offering additional safety briefings to elderly passengers — will help further accelerate the deboarding process.

Children, infants, and pregnant women also introduce unique physical capabilities and behaviors that add another vital layer to evacuation modeling, which the group plans to investigate in their future work.

“We hope these findings help airlines proactively mitigate risks,” Zhang said. “By understanding how passenger distribution affects evacuation, airlines could potentially implement more strategic seating arrangements to optimize safety without compromising operational efficiency.”

###

The article “Effect of elderly passenger distribution on A320 aircraft evacuation under dual-engine fire scenarios” is authored by Xu Zhao, Ying Xia, Chenyang Zhang, Tianchang Meng, Gaobo Yang, Hua Chen, and Yu Zhang. It will appear in AIP Advances on March 31, 2026 (DOI: 10.1063/5.0310405). After that date, it can be accessed at https://doi.org/10.1063/5.0310405.

ABOUT THE JOURNAL

AIP Advances is an open access journal publishing in all areas of physical sciences — applied, theoretical, and experimental. The inclusive scope of AIP Advances makes it an essential outlet for scientists across the physical sciences. See https://pubs.aip.org/aip/adv.

###

---

Journal

AIP Advances

DOI

10.1063/5.0310405 

Article Title

Effect of elderly passenger distribution on A320 aircraft evacuation under dual-engine fire scenarios

Article Publication Date

31-Mar-2026

 

Getting a glimpse of viral dances in the dark in the Sargasso Sea

Study shows time-based patterns that can inform ocean modeling

Ohio State University

FacebookX

LinkedIn
MessageWhatsAppEmail

COLUMBUS, Ohio – In a new study of viral abundance over a short time frame in the Sargasso Sea, researchers found that almost all viruses with cyclical changes in abundance were most active at night – somewhat surprising when the team expected microbial behavior to pick up pace when light was available for photosynthesis.

It turns out the viruses most busy at night were not infecting bacteria that perform photosynthesis, which are among the types of bacteria known to be infected by viruses. Instead, these overnight viral hosts were microbes that focus on consumption of other organic matter because they can’t produce their own food.

The findings reveal another level of complexity of viral interactions with marine bacteria, opening the door to new questions about how these dances in the dark influence ecological services provided by the world’s oceans.

“We still don’t know most of the genes that viruses have and what they do. So it’s elucidating to know that these patterns exist for these viruses and their predicted hosts,” said first study author Alfonso Carrillo, a PhD student in microbiology at The Ohio State University.

The rare look at microbial changes over a short time frame can also be used to inform future models designed to predict how the oceans will respond to warmer, more acidic water – the current conditions in the Sargasso Sea near Bermuda in the Atlantic Ocean.

“To understand how the ocean works as a whole, you can’t exclude the viruses,” Carrillo said. “It’s important to understand how these viruses are behaving, how they’re interacting with host bacteria, and how those interactions change over time. You can’t really make models of how oceans will change unless you know all of these different frameworks.”

The research was published recently in PLOS Biology.

Water samples for the study were taken as part of a long-term initiative called the Bermuda Atlantic Time Series.  

The team sampled water from the surface and in an area called the deep chlorophyll maximum, or DCM, where a lot of microbes that perform photosynthesis are expected to be found. Over the course of 112 hours, they collected surface water every four hours and DCM water every 12 hours.

“We wanted to ask the question, do the viruses change between the depths and do we see any changes with regard to time? We expect it to change because the DCM has higher chlorophyll and there are differences in light, temperature and oxygen compared to surface levels,” Carrillo said.

The composition of the viral communities present in each setting did differ, as expected, and the team then examined viruses that engaged in diel behavior – that is, cyclical changes in abundance within a 24-hour block of time.

Of the over 48,000 virus species collected, almost 3,100 showed diel behavior – and for about 90%, abundance peaked at night instead of during the day.

“This was unexpected because we thought the majority of the viruses that would have this kind of behavior may be performing photosynthesis or targeting bacteria that perform photosynthesis, but that wasn’t the case,” Carrillo said.

Instead, these viruses more active in the dark infected heterotrophic host microbes: those that eat other organisms because they can’t produce their own food.

“That’s interesting to us because it’s something we hadn’t seen before, and it’s something that we can incorporate into future models about how viruses and their hosts might be behaving in the oceans,” he said.

Carrillo works in the lab of Matthew Sullivan, professor of microbiology and civil, environmental and geodetic engineering and director of the Center of Microbiome Science at Ohio State, whose research program focuses on how viruses impact microbiomes in complex ocean, soil and human systems, including pioneering many experimental and bioinformatic approaches to “see” these impacts. Within that context, his lab is investigating how carbon cycling works in the oceans and the role viruses play. 

Better, faster classification of viruses

Though the study of viruses and their functions in the sea, soil and our guts is advancing every day, the extent of what remains unknown about viruses far exceeds what scientists do know. A new analytical tool developed in Sullivan’s lab is helping narrow that gap, using machine learning to establish a rapid classifier of virus samples.

“This tool allows researchers to organize the virosphere, which basically represents all the viruses that we know about,” said first author Benjamin Bolduc, a computational scientist in microbiology. “And that’s actually really important because if you don’t know what viruses are related to other viruses, then it really impacts the kind of knowledge you can glean from whatever area of science that you’re studying.”

The paper was published recently in Nature Biotechnology.

Compared to its earlier versions, the updated tool, called vConTACT3, expands the breadth and depth to which the organization of the virosphere extends in the biological classification of living organisms.

“For decades we’ve been looking at just the species, or just the genera, and that’s important and relevant, but it doesn’t give you any other information,” Bolduc said. The new tool assists researchers in determining relationships at more general levels, such as family, order, class and phylum.

Previous versions of vConTACT also focused only on prokaryotes – archaea and bacteria that lack a nucleus – while vConTACT3 includes viruses that infect eukaryotes, organisms with a membrane-bound nucleus that include all animals, plants and fungi.

Because virus samples collected by researchers are often snippets of these organisms, their genomes are fragmented, and the lack of whole genomes has been a limiting factor in identifying viruses and, by extension, what they do in the environment.

Bolduc applied machine learning techniques at various stages of the development pipeline to identify patterns among genome fragments, which “helps overcome the fact you don’t necessarily need the whole genome anymore in order to accurately classify a virus,” he said.

The team assessed vConTACT3’s performance against reference datasets and large databases of viral genome sequences.

“The tool is a big deal, increasing what virologists use to understand what new virus was discovered, and does so using knowledge-guided AI with some 60 million sensitivity analyses evaluated to fine tune it,” Sullivan said. “It’s also orders of magnitude faster than its predecessors at processing large datasets.”

Both studies were supported by the U.S. National Science Foundation. Carrillo’s work was also supported by the National Institutes of Health; Montgomery County, Maryland; and the University of Maryland. Bolduc’s work was also supported by the German Research Foundation, the Alexander Humboldt Foundation and the Biotechnology and Biological Sciences Research Council.

Sullivan was the senior author of both studies. Co-authors of the PLOS Biology paper included Emily Hageman, Anna Mackey, Kimberley Ndlovu, Funing Tian, Dean Vik, Christine Sun and Richard Pavan of Ohio State; Lauren Chittick of Midwestern University; Naomi Gilbert of Lawrence Livermore National Laboratory; Daniel Muratore of Georgia Institute of Technology; Gary LeCleir and Steven Wilhelm of the University of Tennessee Knoxville; Ho Jang of Korea Virus Research Institute; and Joshua Weitz of the University of Maryland. Co-authors of the Nature Microbiology paper were Olivier Zablocki and Jiarong Guo of Ohio State; Dann Turner of University of the West of England; Ho Jang of Korea Virus Research Institute; Evelien Adriaenssens of the Quadrum Institute; and Bas Dutilh of Freidrich Schiller University.

#

---

Journal

PLOS Biology

DOI

10.1371/journal.pbio.3003474 

Article Title

Sub-daily virus sampling at the Bermuda Atlantic Time Series reveals diel and depth-structured population dynamics without community-level shifts