Arabic Localization of PsychoPy: A Protocol for Expanding Accessibility to Research in Experimental Psychology and Neurocognitive Sciences

Introduction

Recent psychological research has increasingly leveraged open-source software due to its flexibility and accessibility. This facilitates the design and implementation of psychological experiments. Prominent examples include PsychoPy [1, 2]; OpenMaze, an open-source toolbox for virtual navigation experiments [3]; PyControl, a platform integrating open-source hardware and software for behavioral neuroscience experiments [4]; and PsyToolkit, a software package for programming psychological experiments [5].

PsychoPy, in particular, is a popular open-source tool in psychological and neuroscientific research with a reputation for easing the development of behavioral experiments [2]. Developed by Peirce [1], PsychoPy offers multiple Graphical User Interfaces (GUIs), each designed to cater to different user needs and expertise levels. The Builder GUI is a user-friendly, code-free interface where users can design experiments by dragging and dropping components. This mode is ideal for beginners or those without programming skills. In contrast, the Coder GUI provides a script-based environment for advanced users who prefer writing Python code to customize experiments with greater flexibility. The Runner GUI simplifies experiment execution by allowing users to configure settings. For instance, it helps with participant IDs and full-screen modes before running tasks created in either Builder or Coder. Additionally, the Settings/Preferences GUI enables global adjustments, such as screen resolution and audio configurations, ensuring consistency across experiments. Together, these interfaces enhance convenience by accommodating diverse skill levels and experimental requirements, making PsychoPy accessible and adaptable for a wide range of researchers [1, 2]. PsychoPy potential to create accurate visual and auditory stimuli has been very important in neuroscience and psychophysics [11]. Over time, PsychoPy has been developed to provide advanced features, like online support for experiments, and increased usability [2].

However, even with its widespread adoption, at the time this project began, PsychoPy had only been translated into Japanese, limiting its accessibility for non-English researchers. PsychoPy remains unavailable in Arabic, which is a critical gap, given that Arabic is among the world’s five most spoken languages, with over 450 million speakers [6]. This absence restricts equitable access to a key research tool and perpetuates disparities in global scientific participation. While tools like PsyToolkit and OpenMaze support a variety of experiments, none provide Arabic localization—highlighting a systemic gap in open-source psychology software. Zeinoun and her colleagues [7], and Ab. Rahim [8] highlights the absence of Arabic-language scientific tools. This is a subset of a larger issue outlined in the United Nations Educational, Scientific, and Cultural Organization (UNESCO) Science Report [9], explaining the predominance of English in scientific communication and the implications for non-English-proficient researchers.

The urgency underscored by the UNESCO Recommendation on Open Science [9] stresses the imperative necessity of rendering multilingual scientific information freely available, accessible, and reusable to everyone. This principle is essential to democratize worldwide research. Nonetheless, PsychoPy’s English-centric architecture [2] has restricted accessibility for cognitive scientists and students who use Arabic. This led to impeding their capacity to use the platform to its fullest potential and contribute to the research, thus adding to disparities in scientific progress. Solving the localization problems encountered in PsychoPy is essential to enhance inclusivity and support global open science activities for underrepresented linguistically diverse communities [10].

Arabic localization of PsychoPy presents unique technical and linguistic challenges beyond standard software translation. Right-to-left (RTL) script rendering requires specialized GUI adaptation; Arabic diglossia (the coexistence of Modern Standard Arabic and regional colloquials) necessitates careful terminology choices; and psychological terminology often lacks direct Arabic equivalents, requiring contextually appropriate solutions. These challenges are not merely technical—they directly impact usability for 450+ million Arabic speakers and affect the validity of cross-cultural psychological research.

The localization and translation of psychological tools are essential to ensure cross-cultural usability and accessibility across different cultures. Hambleton and Zenisky [12] focus on the importance of strict translation methodology, such as forward and back-translation, to ensure the validity of such tools. Also, Behr and Shishido [13] emphasize the challenges of maintaining consistency in terminology and cultural impact through the translation process. This study aimed at the translation and standardization of the PsychoPy program for the Arabic-speaking population.

The PsychoPy package already supported hooks to identify terms in the software that require translating, such as the menu items and dialog box labels, which can be systematically detected and compiled into portable object (PO) and machine object (MO) files using a package such as Poedit (https://poedit.net/). The architecture was therefore in place to support translation into additional languages.

Given the growing demand for accessible experimental tools among Arabic-speaking researchers and clinicians, this project aimed to translate PsychoPy into Modern Standard Arabic (MSA, or Fuṣḥā), thereby enhancing global participation in psychological science. As part of the engagement initiation launched by the core PsychoPy team at the University of Nottingham and in collaboration with the Department of Psychology at the University of Jordan, the PsychoPy software was translated into the Arabic language. This paper demonstrates the translation process, the challenges encountered, and the implications of this important work for cross-cultural research.

Implementation

Methodology overview

This project aimed to translate the PsychoPy software (version 2023.2.3) into Arabic to make it more accessible and user-friendly for researchers and educators in Arabic-speaking regions. The translation process involved a collaborative effort between three team members (i.e., AA, MA, and AAn). The project encompassed translating 2,067 strings from the original English version while ensuring accuracy, contextual relevance, and cultural appropriateness (See Figure 1).

Figure 1

Quality control

The methodology entailed systematically identifying and translating code strings while preserving their functional integrity and contextual relevance. All translation activities were performed using the PsychoPy platform and documented in detail to maintain transparency and reproducibility. Interface elements and messages were translated with a focus on clarity, accessibility, and user-friendliness, adhering strictly to PsychoPy’s operational requirements. Code strings containing commands, variable names, and functions were intentionally left in their original English format when essential for maintaining software compatibility. In translating instructional texts and user-facing content, the team adhered to best practices as recommended by Van Nes and colleagues [21]. This involved employing contextually appropriate language to convey meaning accurately, rather than relying on literal translations that might distort the intended message. The team remained cognizant of the nuanced differences between English and Arabic, particularly in technical terminology, to mitigate potential semantic shifts. The following subsections detail the specific quality control procedures undertaken.

V. Challenges and Solutions

Key challenges encountered during the project included contextually inaccurate AI translations, version control training, and consistency in terminology. Many AI-generated translations were either overly literal or contextually inaccurate. To address this, manual reviews and adjustments were conducted to refine the translations. Additionally, some team members were initially unfamiliar with GitHub, necessitating training sessions to build proficiency in version control commands and processes. These sessions ensured smooth collaboration and seamless integration of work across the team.

To maintain consistency in terminology, routine meetings and ongoing discussions were held, enabling the team to establish and adhere to standardized terminology throughout the software development process. The translation team utilized collaborative platforms such as Zoom [16] for virtual meetings. All discussions were recorded and stored securely in compliance with the European General Data Protection Regulation (GDPR) [17]. Translations were systematically reviewed and validated by team members to ensure accuracy and consistency. An initial set of code strings and interface elements was translated and subsequently reviewed by an experienced researcher (JP) to establish a robust and consistent coding framework (See Figure 2 for the Arabic GUI of Builder mode in PsychoPy software 2025.1.0 version).

Figure 2

Key challenges specific to Arabic included rendering right-to-left (RTL) script within the interface, managing diglossia—the distinction between Modern Standard Arabic and various dialects—and adapting psychological terminology to fit specific cultural nuances. The primary challenges of this project—contextual inaccurate AI translations, version control complexities, and terminological consistency—highlight critical areas for improvement in collaborative multilingual software development. AI-generated translations often lacked contextual appropriateness, underscoring the ongoing need for human oversight to ensure accuracy [14]. Initial unfamiliarity with GitHub revealed the importance of early training in version control systems to prevent workflow disruptions. Finally, while regular meetings and discussions helped achieve terminological consistency, minor discrepancies may persist, reflecting the inherent challenges of standardizing translations across diverse linguistic contexts [12]. These findings emphasize the value of combining advanced tools with structured collaboration to address such challenges effectively.

1. Leverage Existing Internationalization Infrastructure but Verify Its Limits

PsychoPy’s PO/MO-based i18n framework enabled systematic string extraction and integration without core code changes.

Lesson: Test target-language UI mockups early, including text expansion, RTL directionality, and font rendering, as they often disrupt layouts.

2. Adopt a Rigorous, Multi-Stage Translation Methodology

We applied a forward-backward protocol for conceptual fidelity in technical terms: three independent English-to-MSA forward translations, Back-translation by a fourth bilingual expert, and reconciliation meetings. This ensured semantic equivalence. Domain expertise trumps bilingual fluency alone; pre-validate a glossary of core terms.

Recommendation: Use forward-backward cycles with expert reconciliation for accuracy and clarity.

3. Combine AI-Powered Drafting with Human Expert Review

DeepL via Poedit drafted over 2,000 strings but often missed nuance (e.g., “trial” in experiments). Bilingual cognitive scientists corrected outputs.

Lesson: AI accelerates but requires domain-expert review for technical/contextual strings.

Recommendation: Draft with AI, refine with bilingual scientific human experts.

4. Implement Strong Version Control and Provide Team Training

Git/GitHub managed files and synced with PsychoPy updates. Onboarding resolved non-technical contributors’ initial hurdles.

Recommendation: Adopt Git from day one; train all contributors to prevent conflicts.

5. Establish Communication Channels and Ensure Terminological Consistency

Weekly online meetings and a shared channel unified terms like “stimulus” across parallel translators.

Recommendation: Develop a shared glossary and schedule check-ins upfront.

6. Anticipate and Address Language-Specific Technical Challenges

Arabic’s RTL script caused UI misalignment, text overflow, and font inconsistencies. PsychoPy handled strings; layouts needed developer fixes.

Recommendation: Prototype RTL/complex-script UIs early with developers.

7. Engage Proactively with the Core Development Team

PsychoPy core team input ensured standards compliance, leading to Arabic in v2025.1.0.

Recommendation: Involve developers early for feasibility, integration, and sustainability.

8. Think Beyond the Interface: Support the Broader Ecosystem

Extend to documentation, tutorials, and communities. Our process documentation aids reproducibility.

Recommendation: Translate ecosystem elements; publish methodology (e.g., https://github.com/psychopy/psychopy/commit/96e34c210fb0de3568f16f0ac6f15954b2f1969b).

In summary, to replicate this work: (1) Fork the PsychoPy repository; (2) Use Poedit to extract .po files; (3) Apply forward-back translation with domain experts; (4) Resolve RTL/layout issues in consultation with developers; (5) Submit via PR. Full steps are detailed in the 8-point protocol above.

Limitations and further works

A few factors limit this study’s scope, such as the generalizability limitation to varied Arabic-speaking populations, especially since regional dialectical variations are not comprehensively addressed by prioritizing MSA. Technical limitations persist, particularly incomplete RTL support for dynamic GUI elements (e.g., variable placeholders) and inconsistent font rendering, which may adversely affect user experience. Furthermore, the emphasis on UI and documentation translation ignores ecosystem-wide barriers, such as a lack of Arabic-language technical support forums or training materials, that may inhibit adoption. Ongoing attention to locale-specific conventions (e.g., number formatting, punctuation) remains essential for full i18n compliance.

Future work may focus on developing an ecosystem, including Arabic-language technical support forums or training materials. Further, investigating how semi-automated machine learning pipelines could complement the manual approach and potentially extend localization efforts to other languages. Additionally, integrating a community-based review system for translations could further refine the quality and applicability of localized versions.

The localization of PsychoPy into Arabic represents a crucial step in expanding access to experimental psychology tools. The structured translation methodology, rigorous quality control measures, and collaborative efforts ensured the production of a reliable and culturally appropriate Arabic version. The integration of this translation into the official PsychoPy from release 2025.1.0 onwards will facilitate broader engagement and inclusivity in psychological research, supporting a more diverse global scientific community. Future initiatives should build upon this work by exploring additional language adaptations and refining best practices for software localization in psychological research.

Broader Impacts and Ethical Considerations

This project transcends technical translation. It challenges linguistic bias in experimental design and promotes culturally responsive science. By enabling Arabic-speaking researchers to design and run studies in their native language, we reduce cognitive load, minimize misinterpretation, and foster more authentic participation. The initiative also contributes to Human-Computer Interaction (HCI) research, particularly in RTL interface usability, and aligns with FAIR principles across linguistic contexts. Furthermore, it underscores an ethical responsibility to close the digital and linguistic divide in science. When only English-speaking researchers can fully access advanced tools, global knowledge production remains inequitable.