Jove
Visualize
Contáctanos

Videos de Conceptos Relacionados

Comparing Experimental Results: Student's t-Test01:09

Comparing Experimental Results: Student's t-Test

2.2K
The t-test is a statistical method used to compare the sample mean with a population mean or compare two means from two data sets. The test statistic is calculated from the standard deviation, mean, and number of measurements in the data set at a selected confidence interval and then compared to a table of critical values at this confidence level. If the test statistic is smaller than the critical value, the null hypothesis is accepted. In this case, we state that the difference between the...
2.2K
Binet's Contribution to Measures of Intelligence01:23

Binet's Contribution to Measures of Intelligence

1.4K
Alfred Binet, along with his student Théophile Simon, was tasked by the French Ministry of Education in 1904 to create a method for identifying students who struggled to learn through conventional classroom instruction. This initiative aimed to address overcrowding by placing such students in specialized schools. Binet and Simon developed an intelligence test comprising 30 tasks, ranging from simple commands, like touching one's nose or ear, to more complex tasks, such as drawing...
1.4K
Microsoft Excel: Student's t-Test01:25

Microsoft Excel: Student's t-Test

674
Student's t-test in Microsoft Excel is a statistical method used to compare the means of two groups to determine if they are significantly different from each other. It's commonly used to evaluate hypotheses, such as testing whether a treatment has an effect compared to a control group. Excel provides built-in functions to perform t-tests, making it accessible for users needing to conduct basic statistical analysis.
To conduct a t-test in Excel, use the T.TEST function or the "Data...
674
Introduction to Test of Independence01:21

Introduction to Test of Independence

2.5K
In statistics, the term independence means that one can directly obtain the probability of any event involving both variables by multiplying their individual probabilities. Tests of independence are chi-square tests involving the use of a contingency table of observed (data) values.
The test statistic for a test of independence is similar to that of a goodness-of-fit test:
2.5K
Reliability and Validity01:29

Reliability and Validity

13.2K
Reliability and validity are two important considerations that must be made with any type of data collection. Reliability refers to the ability to consistently produce a given result. In the context of psychological research, this would mean that any instruments or tools used to collect data do so in consistent, reproducible ways.
13.2K
Test Cross01:39

Test Cross

42.4K
Alleles are different forms of the same gene. Humans and other diploid organisms inherit two alleles of every gene, one from each parent.
42.4K

También podría leer

Artículos Relacionados

Artículos vinculados a este trabajo por autores compartidos, revista y gráfico de citas.

Ordenar por
Same author

Words vs. worlds.

Science (New York, N.Y.)·2026
Same author

Quantum 'thermometer' takes temperatures inside living cancer cells.

Nature·2026
Same author

Chats with sycophantic AI make you less kind to others.

Nature·2026
Same author

Is the journal legitimate? Aletheia-Probe can help you decide.

Nature·2026
Same author

Eureka Cam: Movements reveal moment of mathematical discovery.

Scientific American·2025
Same author

Google AI aims to make best-in-class scientific software even better.

Nature·2025
JoVE
x logofacebook logolinkedin logoyoutube logo
ACERCA DE JoVE
Visión GeneralLiderazgoBlogCentro de Ayuda JoVE
AUTORES
Proceso de PublicaciónConsejo EditorialAlcance y PolíticasRevisión por ParesPreguntas FrecuentesEnviar
BIBLIOTECARIOS
TestimoniosSuscripcionesAccesoRecursosConsejo Asesor de BibliotecasPreguntas Frecuentes
INVESTIGACIÓN
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchivo
EDUCACIÓN
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualCentro de Recursos para ProfesoresSitio de Profesores
Términos y Condiciones de Uso
Política de Privacidad
Políticas

Video Experimental Relacionado

Updated: Sep 24, 2025

Author Spotlight: Validation of SICOLE-R for Assessing Cognitive and Reading Skills in Spanish-Speaking Children and Its Role in Personalized Education
09:00

Author Spotlight: Validation of SICOLE-R for Assessing Cognitive and Reading Skills in Spanish-Speaking Children and Its Role in Personalized Education

Published on: August 16, 2024

924

Enseñado para la prueba

Matthew Hutson1

  • 1Matthew Hutson is a journalist in New York City.

Science (New York, N.Y.)
|May 10, 2022
PubMed
Resumen

El software de inteligencia artificial (IA) sobresale en las preguntas complejas de la prueba de CI, pero falla en tareas simples de razonamiento. El desarrollo de puntos de referencia mejorados de IA es crucial para el avance de la inteligencia general artificial.

Área de la Ciencia:

  • Inteligencia artificial
  • Ciencias cognitivas
  • Psicometría

Sus antecedentes:

  • Los sistemas actuales de IA demuestran capacidades avanzadas en dominios específicos, a menudo superando el rendimiento humano.
  • Sin embargo, la IA a menudo exhibe fragilidad, fallando inesperadamente en tareas que requieren sentido común o razonamiento general.
  • Las pruebas estandarizadas de inteligencia, como las pruebas de CI, se utilizan cada vez más para comparar el progreso de la IA.

Objetivo del estudio:

  • Evaluar el rendimiento de los modelos avanzados de IA en un conjunto completo de pruebas de inteligencia psicométrica.
  • Identificar las debilidades específicas y los modos de falla de la IA en las tareas cognitivas.
  • Explorar la utilidad de los puntos de referencia mejorados para la evaluación de la inteligencia artificial general (IAG).

Más Videos Relacionados

Mixed Reality for Education MRE Implementation and Results in Online Classes for Engineering
04:12

Mixed Reality for Education MRE Implementation and Results in Online Classes for Engineering

Published on: June 23, 2023

759
Making Sense of Listening: The IMAP Test Battery
11:25

Making Sense of Listening: The IMAP Test Battery

Published on: October 11, 2010

15.9K

Videos de Experimentos Relacionados

Last Updated: Sep 24, 2025

Author Spotlight: Validation of SICOLE-R for Assessing Cognitive and Reading Skills in Spanish-Speaking Children and Its Role in Personalized Education
09:00

Author Spotlight: Validation of SICOLE-R for Assessing Cognitive and Reading Skills in Spanish-Speaking Children and Its Role in Personalized Education

Published on: August 16, 2024

924
Mixed Reality for Education MRE Implementation and Results in Online Classes for Engineering
04:12

Mixed Reality for Education MRE Implementation and Results in Online Classes for Engineering

Published on: June 23, 2023

759
Making Sense of Listening: The IMAP Test Battery
11:25

Making Sense of Listening: The IMAP Test Battery

Published on: October 11, 2010

15.9K

Principales métodos:

  • Utilizó modelos de IA de última generación, incluidos los grandes modelos de lenguaje (LLM), para responder preguntas de las baterías de pruebas de CI establecidas.
  • Analizó el rendimiento de la IA en diferentes dominios cognitivos, como el razonamiento verbal, el pensamiento abstracto y la visualización espacial.
  • Categorizó los errores de IA para comprender la naturaleza de sus fallas.

Principales resultados:

  • Los modelos de IA lograron altas puntuaciones en muchos elementos de prueba de CI desafiantes, demostrando un reconocimiento de patrones sofisticado y una recuperación de conocimientos.
  • Se observaron fallas significativas en tareas que requieren sentido común básico, razonamiento causal y comprensión del contexto implícito.
  • El análisis de errores reveló una tendencia de la IA a cometer errores ilógicos o sin sentido en problemas aparentemente simples.

Conclusiones:

  • Si bien la IA es prometedora en tareas cognitivas específicas, los modelos actuales carecen de una inteligencia general robusta y un razonamiento de sentido común.
  • Los puntos de referencia existentes pueden sobreestimar las capacidades de la IA debido a su enfoque en tareas intensivas en conocimiento.
  • El desarrollo de puntos de referencia más matizados y desafiantes es esencial para guiar la futura investigación de IA hacia AGI.