Category: 3. Business

TOKYO, Japan ― Renesas Electronics Corporation (TSE:6723), a premier supplier of advanced semiconductor solutions, today introduced the RA8M2 and RA8D2 microcontroller (MCU) groups. Based on a 1 GHz Arm^® Cortex^®-M85 processor with an optional 250 MHz Arm^® Cortex^®-M33 processor, the new MCUs are the latest Renesas offerings to deliver an unmatched 7300 Coremarks of raw compute performance, the industry benchmark for MCUs. The optional Cortex-M33 processor enables efficient system partitioning and task segregation.

Both RA8D2 and RA8M2 devices are ultra-high performance MCUs as part of the second generation of the RA8 Series – the RA8M2 are general-purpose devices, and the RA8D2 MCUs are packed with a variety of high-end graphics peripherals. They are built on the same high-speed, low-power 22-nm ULL process used for the RA8P1 and RA8T2 devices introduced earlier this year. The devices include single and dual core options, and a specialized feature set to address the needs of a broad base of compute intensive applications. They take advantage of the high performance of the Arm Cortex-M85 processor and Arm’s Helium™ technology to offer a significant performance boost for digital signal processor (DSP) and machine learning (ML) implementations.

The RA8M2 and RA8D2 devices offer embedded MRAM that has several advantages over Flash technology – high endurance & data retention, faster writes, no erase needed, and byte addressable with lower leakage and manufacturing costs. SIP options with 4 or 8 MB of external flash in a single package are also available for more demanding applications. Both the RA8M2 and RA8D2 MCUs include Gigabit Ethernet interfaces and a 2-port TSN switch to address industrial networking use cases.

Both of the MCU Groups provide a combination of the high performance of the Cortex-M85 core, together with large memory and a rich peripheral set, making them particularly suitable for a wide range of IoT and industrial use cases. The lower power CM33 core can act as a housekeeping MCU, executing system tasks while the high performance CM85 core stays in sleep mode, to be woken up only as needed for high compute tasks, thus lowering the system power consumption.

“The RA8M2 and RA8D2 complete Renesas’ new generation of RA8 MCUs, purpose-built for the high-performance microcontroller market,” said Daryl Khoo, Vice President of the Embedded Processing Marketing Division at Renesas. “This portfolio empowers Renesas to deliver scalable, secure and AI-enabled embedded processing solutions that accelerate customer innovation and time-to-market across a broad spectrum of industrial, IoT and select automotive applications. Renesas’ commitment to innovation is reflected in the RA8 Series’ ability to address complex processing requirements while maintaining lower power consumption and minimizing total cost of ownership to future-proof customers’ designs.”

RA8D2 Feature Set Optimized for Graphics and HMI Applications

The RA8D2 MCUs provide a plethora of features and functions for graphics and HMI applications:

• High resolution Graphics LCD Controller supports up to 1280×800 displays with both parallel RGB and 2-lane MIPI DSI interfaces
• Two-Dimensional Drawing Engine offloads the graphics rendering tasks from the CPU and supports graphics primitives
• Multiple camera interface options enable camera and vision AI applications,
  • 16-bit camera interface (CEU) with support for image data fetch, processing and format conversion
  • MIPI CSI-2 interface offers a low pin-count interface with 2 lanes, each up to 720Mbps
  • A VIN module performs vertical and horizontal scaling and format and color space conversions of YUV and RGB data inputs received from the MIPI CSI-2 interface
• Audio interfaces such as I2S and PDM support digital microphone inputs for audio and voice AI applications
• Comprehensive graphics solution with industry-leading embedded graphics GUI packages from SEGGER emWin and Microsoft GUIX, integrated into Renesas’ FSP
• Software JPEG decoder optimized for Helium, available with both emWin and GUIX solutions, allows decode of JPEG images with up to 27fps end-to-end graphics performance with Helium acceleration
• Multiple graphics ecosystem partners such as Embedded Wizard, Envox, LVGL and SquareLine Studio are offering solutions that employ RA8D2 using Helium to accelerate graphics functions and JPEG decoding

Key Features of the RA8M2 and RA8D2 Group MCUs

• Core: 1 GHz Arm Cortex-M85 with Helium; Optional 250 MHz Arm Cortex-M33
• Memory: Integrated 1MB high-speed MRAM and 2MB SRAM (including 256KB TCM for the Cortex-M85 and 128KB TCM for the M33). 4MB and 8MB SIP devices coming soon.
• Analog Peripherals: Two 16-bit ADC with 23 analog channels, two 3-channel S/H, 2-channel 12-bit DAC, 4-channel high-speed comparators
• Communications Peripherals: Dual Gigabit Ethernet MAC with DMA, USB2.0 FS Host/Device/OTG, CAN2.0 (1Mbps)/CAN FD (8Mbps), I3C (12.5Mbps), I²C (1Mbps), SPI, SCI, Octal serial peripheral I/F
• Advanced Security: RSIP-E50D Cryptographic engine, robust secure boot with FSBL in immutable storage on-chip, secure debug, secure factory programming, DLM support, tamper protection, DPA/SPA protection,

The new RA8M2 and RA8D2 Group MCUs are supported by Renesas’ Flexible Software Package (FSP). The FSP enables faster application development by providing all the infrastructure software needed, including multiple RTOS, BSP, peripheral drivers, middleware, connectivity, networking, and security stacks as well as reference software to build complex AI, motor control and cloud solutions. It allows customers to integrate their own legacy code and choice of RTOS (FreeRTOS and Azure RTOS) with FSP, thus providing full flexibility in application development. In addition, Zephyr support is now included. Using the FSP will ease migration of existing designs to the new RA8 Series devices.

Winning Combinations

Renesas has combined the new RA8 Group MCUs with numerous compatible devices from its portfolio to offer a wide array of Winning Combinations, including the Smart Glasses and Pet Camera Robot for the RA8M2, and both Ki Wireless Power Transceiver System (Tx) and Ki Wireless Power Receiver System (Rx) for the RA8D2. Winning Combinations are technically vetted system architectures from mutually compatible devices that work together seamlessly to bring an optimized, low-risk design for faster time to market. Renesas offers more than 400 Winning Combinations with a wide range of products from the Renesas portfolio to enable customers to speed up the design process and bring their products to market more quickly. They can be found at renesas.com/win.

Availability

The RA8M2 and RA8D2 Group MCUs are available now, along with the FSP software. The RA8M2 devices are available in 176-pin LQFP, 224-pin and 289-pin BGA packages. The RTK7EKA8M2S00001BE Evaluation Kit is also available. The RA8D2 MCUs are offered in 224-pin and 289-pin BGA packages. The RTK7EKA8D2S01001BE Evaluation Kit supports the RA8D2 devices. Information on all these offerings is available at www.renesas.com/RA8M2 and www.renesas.com/RA8D2.

Renesas MCU Leadership

The world leader in MCUs, Renesas ships more than 3.5 billion units per year, with approximately 50% of shipments serving the automotive industry, and the remainder supporting industrial and Internet of Things applications as well as data center and communications infrastructure. Renesas has the broadest portfolio of 8-, 16- and 32-bit devices, delivering unmatched quality and efficiency with exceptional performance. As a trusted supplier, Renesas has decades of experience designing smart, secure MCUs, backed by a dual-source production model, the industry’s most advanced MCU process technology and a vast network of more than 250 ecosystem partners. For more information about Renesas MCUs, visit renesas.com/MCUs.

About Renesas Electronics Corporation

Renesas Electronics Corporation (TSE: 6723) empowers a safer, smarter and more sustainable future where technology helps make our lives easier. A leading global provider of microcontrollers, Renesas combines our expertise in embedded processing, analog, power and connectivity to deliver complete semiconductor solutions. These Winning Combinations accelerate time to market for automotive, industrial, infrastructure and IoT applications, enabling billions of connected, intelligent devices that enhance the way people work and live. Learn more at renesas.com. Follow us on LinkedIn, Facebook, X, YouTube, and Instagram.

(Remarks). All names of products or services mentioned in this press release are trademarks or registered trademarks of their respective owners.

---

The content in the press release, including, but not limited to, product prices and specifications, is based on the information as of the date indicated on the document, but may be subject to change without prior notice.

 

Mexico City, Mexico, October 22, 2025 – ZTE Corporation (0763.HK / 000063.SZ), a global leading provider of integrated information and communication technology solutions, recently kicked off its 2025 Broadband User Conference in Mexico City, Mexico.

 

This year’s congress themed “AI Boosting Broadband Benefits”, brought together over 400 global ICT industry leaders, top telecom operators, enterprise innovation pioneers, and strategic partners to exchange insights on broadband innovations in the intelligent era.

 

The conference highlighted a series of AI-driven innovations, showcasing how AI is accelerating the evolution of next-generation broadband solutions and driving network intelligent transformation to new heights.

 

In the opening keynote, Xiao Ming, President of ZTE Overseas, emphasized that broadband is evolving from mere connectivity to intelligence, serving as the digital neurons system of modern society. He noted that today’s users expect networks that are faster, smarter, and more caring. Guided by this vision, ZTE is accelerating the shift toward intelligent broadband, empowering operators to enhance user experience, improve operational efficiency, and unlock new growth opportunities.

 

Xiao Ming further outlined ZTE’s strategic roadmap, positioning enhanced network experience as the foundation, optimized user interaction as the catalyst, and expanded scenario value as the vision. Through integrating AI-powered smart home solutions, advanced networking, and intelligent operations, ZTE aims to help global partners achieve new levels of business excellence and sustainable growth.

 

“On the journey from ‘Connectivity Provider’ to ‘Intelligence Creator’, ZTE will continue to collaborate with global partners to co-build a sustainable, intelligent world driven by openness, collaboration, and shared success,” Xiao Ming concluded.

 

A key highlight of the conference was the panel discussion session “Winning the Fiber Race: Are Operators Truly Capturing the Rewards?”, which gathered senior leaders from telecoms, enterprises, and digital ecosystems to explore how Fibre drives growth.

 

Meanwhile, three dedicated exhibition zones showcased ZTE’s latest innovations in AI×Optical Network, AI Ecosystem, and Sports Innovation, offering visitors a comprehensive view of ZTE’s cutting-edge technologies and real-world applications.

 

AI×Optical Network

In the FTTx field, ZTE’s Light PON and Light ODN solutions support smooth evolution, enabling faster and low-cost network deployment with compact and outdoor OLTs and simplified network architecture, meeting the network construction demands of ISP and MSO. AI technologies bring in higher quality and efficiency. For example, the iintelligent ODN network planning tool enables one-click network planning and design, effectively reducing the time and labor, and fiber fingerprint technology provides full visibility and management, achieving fault location accuracy within one meter and cutting fault location time by over 90%.

 

In terms of optical transport networks, focusing on “OTN for AI” and “AI for OTN”, ZTE proposes the HI-OTN solution for high performance high intelligence. With single-wavelength 1.6T and 12THz full-band technologies that enable single-fiber 100T transmission, it provides an ultra-broadband and ultra-fast optical network foundation for AI services. Moreover, the AI-powered optical network health assessment and precise optical module failure warning functions shorten the fault maintenance time by 30%. By integrating AI-powered root-cause analysis for alarms, ZTE has boosted troubleshooting efficiency by 20%, significantly accelerating optical network O&M.

 

For IP networks, the AI HI-IPNet solution accelerates the monetization of IP networks. With AI-Native 400GE/800GE high-performance routers, it helps operators build networks with optimal TCO. The end-to-end millisecond-level low-latency slicing supports the development of high-value services such as private lines, cloud gaming, and real-time interaction, boosting revenue growth for operators. Furthermore, AIOps enables proactive O&M and minute-level precise fault diagnosis, ensuring highly efficient and stable network operation.

 

Regarding AI home, ZTE leads a new era of smart life. For home connectivity, ZTE AI Wi-Fi 7 and Gigabit FTTR are equipped with innovations such as AI-powered intelligent antenna, seamless roaming, and acceleration engine. This enables operators to deliver tailored experience by providing scenario-based packages based on thriving ecosystem. As to smartness, the AI media terminals have transformed a single TV box to a full-scenario integrated terminal. It reduces the cost by 30% and increases the ARPU by 20%, redefining the new large screen ecosystem. At the same time, the triple value engine of “smart viewing, smart storage, and smart connectivity” creates a new carrier-class paradigm of “security as a service”. Furthermore, ZTE’s AI Smart View won the “Most Innovative Smart Home Experience” award at Network X 2025.

 

Focusing on “experience as a service”, ZTE AI home provides every family with ultimate full-scenario smart experience through AI innovations and scenario-based services.

 

AI Ecosystem

As AI applications demand scalable, high-performance computing, ZTE presented its end-to-end intelligent computing data center solutions, featuring modular, prefabricated intelligent computing containers with liquid cooling, cutting delivery time by 40%. In power distribution field, intelligent bus bar pooling is adopted to support a wide range of cabinet power densities. In response to the uncertainty of future businesses, it provides a compatible and co-construction solution. 

 

Meanwhile, by combining AI-enabled intelligent management system perception and autonomous learning, it achieves optimal energy efficiency during the operation, meeting the requirements for efficient and flexible deployment.

 

ZTE also showcased its full-stack intelligent computing solution, featuring a comprehensive range of general and AI servers, including the industry-focused AiCube all-in-one appliance, to build a high-performance computing foundation. On the software side, it offers a navigational low-code AI platform and Co-Sight Agent Factory, which lower the barrier for AI model training and inference, enable pipeline-based, minute-level AI app development，allowing small and medium-sized enterprises to build their own AI capabilities amid the wave of digitalization.

 

The ultimate enabler of intelligent computing is energy. ZTE Digital Energy provides the energy foundation for AI computing and communication networks. The renewable energy business exhibition area showcases solutions for PV, energy storage, and the integration of PV with storage, offering smart and clean energy to global operators and enterprise clients. It provides full-scenario, end-to-end EPC capabilities and qualifications for renewable energy projects. The energy storage system features triple-layer protection for ultimate safety. The newly launched “Energy Matrix” integrates storage and generation, supports intelligent hybrid use and multi-energy convergence, and includes intelligent modular power units with over 98% efficiency and a high power density of 70.2W/in³. It supports single-circuit load power-off and metering. Combined with the “One Site, One Strategy” energy operation solution, it reduces solar energy waste by 20%, saves 50% in fuel consumption, and cuts electricity usage by 25%, leading the way in energy digitalization.

 

Sports Innovation

In anticipation of the FIFA World Cup 2026 coming to the Americas, ZTE unveiled the “One Day for Football Fans” experience zone, demonstrating how its technologies elevate fan engagement—from ultra-fast live streaming and immersive multi-screen viewing to seamless connectivity across stadiums and fan zones.

 

ZTE’s FTTR solutions extend ultra-gigabit optical connectivity to hotels and retail stores, enhancing both operational efficiency and user experience. AI-powered cameras and monitoring systems further improve safety and service quality.

 

For large venues, the AI-Optical Stadium solution deliver all-optical 10 Gbps FTTR-B networks, significantly shortening the service provisioning time while increasing the bandwidth.

 

The AI All-Scenario Live Broadcast solution, based on the Cloud Premium Video Platform and hybrid cloud architecture, serving over 150 operators in more than 40 countries, supports 4K/8K ultra-HD playback, cross-device continuity, and AI-enhanced operations, redefining the viewing experience.

 

Throughout the conference, ZTE and its partners reaffirmed their commitment to digital intelligence co-construction across Latin America.

Canon’s RF/EF lens production exceeds 170 million units, extending its world record in interchangeable lens production

TOKYO, October 22, 2025—Canon Inc. announced today that, in October 2025, Canon reached a historic milestone of producing a cumulative total of 170 million RF and EF interchangeable lenses for its EOS series, extending its world record¹ for the highest number of interchangeable camera lenses ever produced.

The EF lens was introduced in 1987 as the dedicated lens system for Canon’s EOS autofocus single-lens reflex film camera, debuting simultaneously with the EOS system itself. Since their inception, EF lenses have led the industry by incorporating a series of world-first² technologies, including the Ultrasonic Motor (USM), Image Stabilizer (IS) technology, and a multilayered Diffractive Optical (DO) element, and have undergone numerous evolutions.

In 2018, Canon launched the RF lens series, designed for the EOS R mirrorless camera system, which features a large aperture, short back focus, and high-speed communication system to deliver even higher image quality. The RF and EF lens series lineup now includes a total of 108 models³, covering a wide range of focal lengths⁴ from ultra-wide 10mm to super-telephoto 1200mm. The series also includes the world’s first VR lens⁵, lenses with built-in power zoom suited for video shooting, and even those compatible with power zoom adapters—expanding the scope of creative expression and meeting the diverse needs of users for both still photography and video.

EF lens production began at Canon’s Utsunomiya Plant in 1987. Since then, both EF and RF lenses have gained strong support from a wide range of users—from beginners to professionals—leading to steady growth in production volume. Today, Canon manufactures lenses at five sites: Utsunomiya Plant; Canon Inc., Taiwan; Canon Opto (Malaysia) Sdn. Bhd.; Oita Canon Inc.; and Miyazaki Canon Inc. Milestones include 10 million units produced by 1995 and 50 million by 2009. Then in 2014, Canon became the first company in the world to reach 100 million interchangeable camera lenses produced. In October 2025, the company reached 170 million units, leading to the achievement of this world record. The 170 millionth lens produced was the RF70-200mm F2.8 L IS USM Z.

Canon has maintained the No.1⁶ global market share for digital interchangeable-lens cameras for 22 consecutive years since 2003. Moving forward, Canon will continue to refine its proprietary imaging technologies and further strengthen and expand its lens lineup, pioneering new imaging possibilities and contributing to the continued evolution of photographic and video culture.

• 1

  Includes EF, EF-S, EF-M, EF Cinema, RF, RF-S, and RF Cinema lenses and extenders. As of October 21, 2025 (according to a survey by Canon)

• 2

  Among SLR cameras (according to a survey by Canon)

• 3

  Number of products sold as of October 22, 2025 (including extenders). The number of lens models for sale is different according to
  market figures.

• 4

  Focal length is 5.2mm to 1200mm when including VR lenses

• 5

  An interchangeable digital camera lens that enables VR footage with a single camera. Among interchangeable lens digital cameras released as of October 5, 2021 (according to a survey by Canon)

• 6

  Refers to unit share (according to a survey by Canon)

Highlights in the development of the RF/EF Lens Series

The EF lens, which was introduced alongside EOS in March 1987, has adopted a variety of world-first technologies, including Image Stabilizer (IS) technology, featured in the EF75-300mm f/4-5.6 IS USM released in 1995; a multilayered Diffractive Optical (DO) element, used in the EF400mm f/4 DO IS USM launched in 2001; and Subwavelength Structure Coating (SWC)⁷, applied to the EF24mm f/1.4L II USM released in 2008.

In 2021, Canon launched the EOS VR System, a VR video system consisting of a mirrorless camera⁸, dedicated lens, and PC software, thereby creating a 3D 180° VR video through an interchangeable lens camera. In 2024, Canon began rolling out a new series of hybrid lenses equipped with iris rings, designed to meet the needs of both still photography and professional video production. For zoom lenses, the company has also launched RF24-105mm F2.8 L IS USM Z and RF70-200mm F2.8 L IS USM Z which are compatible with power zoom adapters. For single focus lenses, the company released F1.4 L hybrid prime lens series that unified the size and ring and button position across models. In September 2025, Canon launched RF85mm F1.4 L VCM, the fifth model in this series, demonstrating that it can meet demands in line with the changing times.

• 7

  A special coating with advanced anti-reflective properties

• 8

  For applicable cameras, please visit the official Canon website.

Gold drew attention during LME Week, with prices up by around 60% this year. The rally has been driven by uncertainties over global trade, heightened geopolitical tensions, US fiscal stability and the Fed’s independence. The start of the Fed’s easing cycle also boosted gold, which doesn’t pay any interest. The rally has been driven by physical buying, with central banks and private investors accumulating gold at record volumes.

But after a weeks-long rally that saw the precious metal hitting successive record highs, gold slid the most in 12 years this week. This signalled that some momentum might have been stretched. Gold was dragged down by a combination of factors, including profit-taking across precious metals, easing seasonal demand from Diwali, positive trade talks between China and the US, uncertainty over investor positions amid the US government shutdown, and a stronger dollar. The pullback underscores the risk that the rally might have moved ahead of underlying fundamentals.

But despite this sharp pullback, gold’s outlook remains constructive, underpinned by macro uncertainty and diversification demand.

The shift in central banks’ purchases has been structural, with the pace of buying doubling in 2022 following Russia’s invasion of Ukraine. Central banks’ appetite for gold is driven by concerns from countries about Russian-style sanctions on their foreign assets, as well as shifting strategies on currency reserves. The top year-to-date buyer has been the National Bank of Poland, and it just announced it aims to increase its reserves from 21% to 30%.

ETFs have been another powerful force behind gold’s record-breaking rally this year, with holdings surging in recent weeks. In fact, gold ETFs have added as much gold in September alone as central banks did during the first quarter of this year combined, according to the World Gold Council. With ETF holdings still shy of a peak hit in 2022, there could be room for further increases.

The downside should be limited, supported by geopolitical concerns, sustained central bank demand and expectations of further monetary easing, although near-term volatility may persist. For now, gold’s pullback looks like a healthy correction within a still-positive trend.

Gold and silver wavered, after suffering their steepest selloffs in years on Tuesday as concern their dizzying rallies in recent weeks had left them overvalued.

Spot gold traded near $4,140 an ounce after tumbling as much as 6.3% in the previous session, the biggest intraday drop in more than a dozen years. Silver edged higher after being down 8.7% at one point on Tuesday. The slumps came after technical indicators showed scorching rallies for both metals were likely overstretched.

Ground-breaking design, outstanding performance

As the first-ever three-row SUV in the IONIQ lineup, the IONIQ 9 left a lasting impression during its global debut in Los Angeles. Guests and members of the international press were immediately impressed by the IONIQ 9’s aerodynamic design, commanding presence, and outstanding electric performance.

The IONIQ 9 breaks boundaries in many ways: exterior design elements, such as its sleek silhouette, tapered boat-tail form, and smooth, single-curved roofline, demonstrate a new aerodynamic architecture and bold design step. But where the concept of bringing people together is most on display is IONIQ 9’s cozy and interaction-enhancing interior. Once inside, both drivers and passengers are greeted with a roomy, harmonious space that feels more like a luxurious lounge than a car cockpit. Soothing colors, innovative materials, and ride-enhancing technology come together to create a unique atmosphere that people want to interact in.

Imagine you’re at the grocery store, standing before a selection of snacks. Seemingly without thinking, you skip over the rice crackers to pick out a bag of chips.

These types of choices are called dietary decisions. It’s how we consider many different aspects of a food – including tastiness, healthiness and price – in order to decide what to buy and what to eat.

It’s not well understood how our brains use all these different bits of information when making food choices. When does information about each aspect of the food become available to our brains to consider? That’s what we set out to investigate.

In our new paper published in the journal Appetite, we show how, just hundreds of milliseconds after we have seen a food, many different attributes are reflected in brain activity. This happens extremely fast, long before a person can consciously decide whether or not to buy or eat the food.

Peering inside the brain

Naturally, how fast our brains process the different aspects of foods will affect our dietary decisions.

For example, studies have reported that we may process how tasty we find a food more quickly than how healthy it is. This quirk can bias our choices toward foods that taste better over those that are healthier. Junk foods – tasty but not necessarily good for us – have an edge here.

To investigate how quickly we process different aspects of foods, we used electroencephalography, a method that allows us to record electrical brain activity with millisecond precision.

We recorded people’s brain activity while showing them images of various foods, such as snack items, meats, fruits and sweets. We also asked people to rate each food on many different aspects, such as healthiness, tastiness, calorie content, familiarity, and how much they would like to eat the food.

We then used machine learning techniques to compare patterns of brain activity (how different the brain responses were to different food items) with the patterns of ratings (how differently those foods were rated).

This allowed us to test whether foods that had the largest differences in ratings also had the largest differences in brain activity. In other words – was information about food attributes actually reflected in people’s brain activity?

As it turned out, it was.

Information about different aspects of foods, such as healthiness, calorie content and familiarity, were reflected in the brain activity as early as 200 milliseconds after the food image was presented on the screen.

These rapid brain responses occurred before people could be consciously aware of the food they were seeing. Other aspects of foods, such as tastiness and willingness to eat the food, were reflected in the brain activity slightly later.

Choosing before choosing

These findings suggest that various aspects of foods may grab our attention early and help guide our dietary decisions. The brain assesses many different aspects of foods automatically and with similar timing, shaping our food choices before we’re even aware of them.

Surprisingly, we found that the healthiness of foods was represented in the brain activity earlier than tastiness. While this contradicted some previous findings, our machine learning techniques may have been more sensitive to detect subtle patterns of brain activity associated with each attribute.

There were also similarities in the way people judged different aspects of a food. For example, foods that were less familiar were also rated as being less tasty.

From these patterns of similarity, we identified two key food dimensions that may be particularly important when our brains evaluate foods. The first one is the “processed” dimension: how natural or processed a food is. The second is the “appetising” dimension, which taps into how tasty and familiar we find a food.

Both were reflected in patterns of brain activity very rapidly, about 200ms after seeing a food item.

There’s more than the eye can see

Our findings are most relevant to situations where we only rely on the visual features of foods, such as when ordering groceries or meals online, or using a picture menu at a restaurant. They shed light on how people make snap judgements at the supermarket or on food delivery apps.

Our brain imaging approach can also be used to test if certain strategies, such as deliberately focusing on the healthiness of foods, might change how foods are rapidly appraised and help us improve our choices.

While we used food images in this study, other senses are also important for dietary decisions. Smelling a mango or hearing the sizzle of a frying burger patty are likely processed quickly by the brain as well.

The next step will be to look into these other sensory features of foods, to see how the brain processes not just images of food, but the real deal when placed in front of us.